#11 - Jason Dreyzehner: 2026 CHIPS OP_EVAL, P2S, Loops and TX5 - Fiendish & Friends

Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
Good day all you Liberty lovers and welcome to Fiendish and Friends episode 11 when we have a legend of the Bitcoin Cash

(00:08):
ecosystem
Jason Drazener joining us to talk about his proposed chips and developments including and I've got to take a big breath right now
To get this all out. OP eval, pay2script, loops as well. Not only that but the fresh of the press proposed chip
transaction version 5 and this includes many things including

(00:29):
zero overhead covenants, detached signatures, comprehensive malleability protection, efficient covenant UTXO
recycling, fractional satashis and fractional cash tokens
Wow, someone's been a busy beaver. Hi Jason, welcome to Fiendish and Friends. It's a privilege to have you join us today
Thank you for the intro. I'll try not to let it go to my head

(00:54):
Yeah, to be honest, it's been it's been great seeing this because you know, you just locked in big ints and
The virtual machine limits upgrade and
Pretty much just a few weeks after that you come in and
With some new chips, so no pause for the wicked

(01:15):
By the way, I was actually looking at my notes the last time we were on a podcast together was way back in
2022 that was when we were discussing the cash tokens upgrade
Which was yeah, of course another one of your chips
I guess everyone in the Bitcoin Cash ecosystem is well aware of cash tokens now

(01:36):
Yeah, so anyway before we start delving into the weeds as this is your first time on Finnish and Friends and to ensure that everyone
Out there knows exactly who you are. It'd be great to have an introduction
So Jason, maybe you could just tell everyone who are you what you do and why all this incredible work towards Bitcoin Cash

(01:57):
Thanks. Yeah, I've been working on I came to Bitcoin for the freedom
I've been working on you know, professionally on Bitcoin Cash since 2013
I was at Bitpay from 2013 to 2019 and I've been working on
protocol improvements
Necessary to make some
Exciting

(02:19):
On-chain applications possible since then
Yeah, cool
So you've been around for a long time
It's been a while. It's been a while. Yeah
Okay
Yeah
And I said so cash tokens was the last time we did a podcast together on the space together
And this was one of the more

(02:40):
Hyped upgrades of BCH's history. So just as a reflection to begin with, you know
How have you personally found the results of cash tokens on the ecosystem since the upgrade in?
2023
Yeah, it's going well, there's a ton of work to do and

(03:01):
in a lot of ways
ketchup to do with
many of the other
Popular layer one networks that that in some cases were able to launch with token support and many of the
Applications that we consider to be
Very common in the space
We're designed, you know, for example on aetherium because aetherium was the only player in town to do

(03:25):
Interesting on chain protocols for a while
And as we sort of discovered over the past few years
That was a result almost entirely of some
some rather silly virtual machine limits that that got
reduced in 2010 and
Essentially, no work. No progress was made on them until very recently

(03:49):
so
cash tokens was a was a good way to make a lot of
Financial applications various things like that
So we can express a lot of financial applications in in very short transactions and they're very quick to validate
And they're useful for for representing a lot of things

(04:11):
But then there are also some other virtual machine
Limits that made it very difficult to do
Even pretty trivial computations like like doing the math for a market making algorithm just calculating based on how much
BCH you have and how many tokens you have
What the what the new price should be? That was that was difficult to do even though

(04:32):
You know, that's microseconds on on on a processor right or significantly less than that. I'm
using
So getting vm limits
Fixed up this year
Really opens up the space a lot for some of the other issues that have been sort of on hold since

(04:52):
since 2010 in a lot of ways, so you know satoshi's original virtual machine
Satoshi's original bitcoin system had a very very capable virtual machine the script system
that
Essentially right out of the gate would have supported anything you've seen on any cryptocurrency for sure

(05:15):
Its issues were that it was a very academic approach to the
to the problem and he had essentially just missed a
very obvious
set of exploits
But he he by in taking a very academic approach. He basically tried to make the virtual machine
Work without any limits by making it not Turing complete

(05:40):
And so he didn't add any limits and he kind of told himself that that was sufficient because it's not Turing complete
But of course the real world operates on on computers that that exist in reality
And those computers have limited things like limited memory limited limited compute
So, you know, he he discovered it very

(06:01):
very suddenly like more than a year after the the
Protocol had been launched and was in production like people were using bitcoin
Just transactions back and forth to each other
They were essentially for about a year and a half before on the bitcoin talk forums
Gavin asks. Hey, what's this script thing?
And you can see that the wheels turning very quickly you see like within eight hours

(06:24):
It's really clear that satoshi like hadn't looked at it in a little while and he's on this
He's on this protocol hardening kick and and he looks at the script system again, and he's like, oh god
There's a lot we need to fix here
And he was lucky at that point that it wasn't really used very much because I guess if he were to release something in that state
Today as it was it's just gonna be oh, yeah

(06:44):
Oh for sure. Yeah, I mean like it was possible
I mean there were zero days to take anyone's money from any address like the
You you could just put op return in and get anybody's money out of any address without providing a signature
I mean, there's just a lot of good. Yeah
That's a good basis of a money
Just a good basis of a money
That sounds good. Yeah, that's a good basis of a money. Yeah

(07:09):
Uh in some ways it's a it's an indicator that whoever was working on the script system
At least was a single was probably a single brain because they didn't talk about it with with another person very much
Or they probably would have between each other like, you know penetrated some of the issues here
Just that's why she was one person then from from your perspective at least whoever was working on the script system
Script system was definitely one person and they didn't explain it to anybody else

(07:32):
in the details
Yeah, so um
Or or they didn't explain it to anybody else who was who was technically in the details
um the uh
But yeah, so um
Then you know fast forward a number of years. Um satoshi or i'll say, you know, so in 2010 he he uh, I guess is it 2011
Wow. Yeah, uh, no, it's 2010. Yeah 2010 because he disappears at the end of 2010. Um

(07:56):
Um, that's right
I'm not fresh for sure on this stuff. Um, but uh
He he then rapidly
Um makes a bunch of patches
um
The first of you know, the first is like just a massive overhaul and then another massive overhaul comes like literally three days later

(08:21):
Um, and then like another week later, there's some other massive changes and you know, he's disabling stuff and cutting stuff and setting magic
Magic numbers for various constant limits
um, just absolutely shredding it um
And over the I think there are like six or seven, you know non-trivial consensus changes, uh, he schedules the first soft fork

(08:44):
Um, and and then once it's done he he cleans it up
Um by removing the logic from before because it's not necessary. It applies right over it. Um
the uh
There's uh, he adds the op-nop codes. Um
The 201 opcode limit there's a lot a number of things there and then um, and then he disappears at the end of the year

(09:05):
and and all all
Progress and and in a lot of ways even interest in this mostly mostly fades. Um, it pops back up a little bit
Uh again in 2020 uh 2012 with people talking about op eval which eventually became
uh
Paid a script hash

(09:26):
But for the most part people were interested in use cases
essentially that let people
Uh give out addresses that had multi-signature conditions
and so um
Being able to support multi-signature wallets and coincidentally that is when I started working on bitcoin cash
And and I worked at bitpay on the the first multi-signature wallet

(09:48):
Copay, uh, I I led the design of of copay at that time
um
Yeah, being one of the only designers
With a with a team of a few designers also, um
uh
then the
Multi-sig uh multi-sig stuff

(10:09):
finally
Being possible after pay, uh p2sh, um probably attracted some more interest to the script system
but um then the block size, uh
Block size war essentially stopped all development in this in this
field um
until until 2017
Uh 2018 we got a lot of op codes reintroduced. Um, uh

(10:34):
Uh, there's probably a few other yeah, then there's a check check data sig happened in 2019
Um 2022 we got introspection 2023. Um cash tokens, uh finally
opened back up
um
general computation
on on the virtual machine though with very very

(10:56):
Low limits and so you had to do quite a few workarounds to get anything to get a lot of useful stuff done
Um, there are a lot of things you you could already do even just with cash tokens
Um, because you can express things pretty efficiently for some financial applications
um
but if you start having to do serious math you you have to break out your computations into multiple inputs and

(11:18):
essentially, that's where things get purely theoretical because
very few people
Even really comprehend how to break apart the problem not
Uh certainly not enough to uh build, um
To build, you know user facing real world usable applications using it
Yeah, exactly. I guess this is where the the big ink comes in for the for the upgrade that's locked this may

(11:41):
Yes, and and the virtual machine limits so, you know, uh
This is the virtual vm limits was a was a huge unlock
That was that was preventing
um
a lot of otherwise quite simple improvements to to our
our virtual machine how we can express programs, um simply because

(12:03):
uh
Before we we basically didn't have limits before except for some magic constants that applied at different places
Um, and we're we're relative based on uh, the encoding format
So our virtual machine limits only
Uh were only implicitly defined
And and the encoding format of the transaction was an important part of what the actual limits were

(12:30):
um, and it's kind of insane to even describe it that way like uh
To to know what the worst case performance of the virtual machine was with maximum size blocks producing the maximum compute transactions
Um, you had to account for how many bytes they would take to encode
And then apply your magic constants in the right spots to figure out what the actual limits were

(12:52):
And so that was that was the work for that for last year was like just just
Observing our system sufficiently to see where all of our limits were and then putting explicit limits in those spots
Um, and that was uh, that was a just a gargantuan task that um, that is thank god finally done

(13:13):
Thank god finally done. Um, and so there are a lot of other things now that um
With with explicit limits now, there there's no program that that can be defined in in
In bitcoin caches vm bytecode format. There's no program that can be defined that way without
um
That is not covered by the explicit limits and and it's it's very obviously and directly applied

(13:38):
Um, there's there's nothing you can do that doesn't exhaust some limit. Um
even pushing uh pushing data which
Um, there there are a number of things that you know, when you first look at the problem
It's really easy to be like, okay, we don't need any limits over that
Um, but if you if you don't do it very comprehensively you start to run into issues the moment you start to talk about like

(14:00):
Okay, can we make our fourth?
Uh, our you know our fourth the vm bytecode language. Can we make it do loops?
Can we make a do function application?
Um, can we can we reallow pay to script instead of just pay to script hash?
and a number of other things and the the vm limits chip talks at length about all of the

(14:21):
um, essentially every other proposal, um
About which we knew that has ever that has ever been made
We wanted to make sure that the that the limit the explicit limits we we had
Uh identified
um
Would also work even if we did need to make
You know a small change here that allowed control flow to loop back over one part of the program or or if we added switch

(14:48):
Statements or if we did obi val or things like that
um
so
Yes, the virtual machine limit upgrade for the 2025 upgrade
Makes the the previously implicit limits very explicit and it applies them
Completely comprehensively to every byte you can possibly execute in the virtual machine

(15:08):
Um, and that's the critical critical thing to understand. So, um
Once you have vm limits, let's just get back to that point because yeah
I'm thinking for people that are listening if it wasn't explicitly limited what what what's the outcome?
Like what what's that? Why is that good or why is the opposite of that bad or dangerous?
um, so the

(15:28):
Um, we know that the the state it was in before and you know by definition it wasn't dangerous because that's how it was working
So we know that that was safe, right?
Um, at least so I wouldn't yeah, but but the question is like what can you change while?
While like what can you what kind of improvements can you make without?
Totally deleting that safety guarantee, right?

(15:51):
And obi val is the is the the oldest most most ancient example
Um, you you cannot make any changes i'll say even
This this may maybe will be um
A good indication of just how delicate it was before. I don't I don't think many people
uh really understood that

(16:14):
Changing the minimum transaction size amounted to a 2x
almost 2x
Increase in the worst case performance of the virtual machine
Because the limits were implicit
Um, so, you know the worst case attack before was 100 100 byte transactions

(16:35):
And and the new worst case was 65 byte transactions. And so you get um, and you get an almost 50%
increase in the worst case
Standard, uh, so, you know, this is also in the mempool not just like a miner has to mine an attack block. This is this was
you know
Yeah
In place in on the live network

(16:57):
um
So implicit limits are just a terrible idea. Um
And they they they bleed and and
leak out their their problems over every other
possible improvement to the protocol, um
And they make it really hard to even reason about

(17:17):
uh
What happens if you add support for loops or what happens if you add support for for function definition?
um
Which op eval is a is a way to do function definition. Um
So things that you come to expect things that you you would expect of any programming language. We don't have

(17:37):
because of essentially a vestige of
satoshi's original
Um kind of half-baked approach to limits
And when he ripped all of that out, he he didn't remove this comment at the top of the script interpreter that said
Uh, the script system has no loops
Even though that was no longer his limit strategy. It was not relevant at all to the code below it

(17:59):
But he didn't delete it
So it like misled this this whole maybe, you know, six six to nine years of new developers who who read that who read that code?
Just accepted as I read that comment just accepted it kind of you know
without without critical thought
and and it defined the uh

(18:20):
What people
Came to understand. Yeah came to believe about the the script system, you know being that that people believed that being
Not Turing complete was an important property. Not not just that like being you know
Uh, the the whole discussion about Turing completely was just completely useless

(18:40):
Completely a red hair. It did not matter at all whether or not it was Turing complete
all that mattered is what actually happens when when you validate a transaction and
whether or not you're the the encoding you're using happens to be
Turing complete or if it if it terminates because you have limits somewhere else
And and you just expressed it less efficiently

(19:00):
Um, so either one is the same result. Um, you know, it's either
You either can express or not a use case and it either fails earlier or it doesn't
but uh, anyways
Once we got but getting past the the the vm limits

(19:21):
Upgrade for 2025 it's now possible for us to to look at
some of these
You know big issues for contract authors like right now
We cannot reliably say
Check all of the input values
And add them together
It's not a reliable way to say that that you have to

(19:43):
You have to instead of like saying hey loop through all the inputs and add them together. You have to say
If there's one input add the value of input one if there are two inputs add the value of input one and two if there are
Three inputs add the input a value one two and three and I i'm not kidding like it is it is truly absurd
Um, just a joke that you you wouldn't even consider right?

(20:06):
You know writing in a programming language that looked like this if it wasn't if it wasn't bitcoin or bitcoin cash
um, but you know, we've we've had kind of
a decade of of kind of stagnation on the development there, um
and
Uh, people were really making do and we're we are still making do a lot of contracts and bitcoin cash right now

(20:28):
We're still making do without loops
And uh, and it gets there are other cases where um
So that's the case for loops for op eval
um the the
We can use the very same case if you need to um
Take uh take the average
of

(20:48):
Of two numbers or something, uh and and add it to each input. I mean like just just some sort of
Run some sort of mathematical function on each input
well now you need to if
If there's one input check the input value here and then specify all your code for the mathematical function
And if there are two inputs grab the input one and grab input two and specify all the mathematical code for input one

(21:14):
And specify all the mathematical code for input two and add them together. Whatever. Um, if there are three inputs
Do it the same do the same thing for input one input two input three
And also you have copied and pasted your your mathematical code six times now, right?
It's factorial increase in the size of the program and that's just for one function
Just for one function application
Now if you start getting like if you do a function that needs to call another function

(21:37):
Even if it's just a little bit of like just some very simple math
Um, you are you are duplicating that interior function, you know six times the outside
uh

(22:04):
Testing you still there you go. You're back
Yeah, my time limit, uh, my time limit went off for uh for x
That's so funny
I actually I do you know what just before this because last time last time I did the same thing
I was kicked out because the time limit went off. I have a strict 15 minute time limit of x today

(22:24):
But i'm always skipping, you know, but but but it's to try and control the uh, the x usage but yeah, sorry, please carry on
um
so that um there again so function application and loops are both like
I mean you would you would laugh if somebody tried to introduce you to a programming language that didn't allow you to specify
A subroutine as a function and call it multiple times

(22:47):
Or or didn't allow you to like loop over the number of inputs and check them for something
These are very very basic things
and
And mind you our our virtual machine is is based on the fourth programming language
So the constructs we're talking about are are
You know 50 years old or more

(23:08):
the the op the the loops
Proposal is proposing one of the loop constructions that is
common to most fourth implementations op begin op until
It's uh, I think from the 70s, um at 1970s and um, it's

(23:30):
very
Standard you can actually ask a large language model to to write an implementation of something and then be like oh
A bitcoin cache added op begin op until fix this and it will actually give you the correct results
Because it's look it's trained on lots of uh, lots of fourth code
um, so so it's useful to have uh, you know to just

(23:51):
Add the thing that most fourths use. Um, so that's the loops proposal
Um, the op eval is a a very similar thing
There are different ways you can you can add support for function definition
If you're curious, there's some discussion in the bitcoin cache research forum about uh, kind of the different

(24:14):
Approaches we could take and what their trade-offs are, uh, but in in short, um
op eval is
sort of the the very simplest thing we could possibly do that we can prove
has um a set of use cases for which it will always be the most optimal, um, and so it's it's

(24:38):
uh, it's very easy to and the reason I
I'm, you know interested in it versus like trying to
Bike shed out a specific way to define functions and let users define unspecified op codes or something like that like
um, there there are there are large bike set bike sheds that could be painted but um
we don't necessarily need to because

(25:00):
Uh, we can prove beyond the shadow of a doubt that op eval is the most optimal way
To to do the the the thing that it that it does
For a very large set of use cases for the first application and for the last application of any function
um
so uh
And let me go back because there's probably an unstated, uh an unstated thing there that

(25:26):
When when looking at at protocol upgrades, we we always want to find ways to make the protocol just timeless
Like the protocol should not should not be changed to make something a little bit more convenient right now
Even if we might find a better solution later
Like that is that is not the the level of rigor required for a protocol change
If our grandchildren are going to have to deal with the with the mistakes we make now

(25:50):
Then we want to be absolutely sure that to the very best of our ability everything we every change we possibly make
Will be perfect
As far as we can see into the future
um, yeah, definitely and this is
This is the critical thing right because you know, I i'm very well aware of your vision of bitcoin cash
We'll go into that later

(26:10):
but um, you know really looking into uh
Uh setting this network up to be money for the world so the world reserve a currency that every single person on the planet can use daily
Uh as cash, um, and of course something like that doesn't happen overnight
It takes time to snowball a certain time. It can hit critical mass and explode

(26:31):
That's true
But still if you're building something that's designed for the world, this isn't something that's you know to achieve that
You're not going to be replacing it then a few weeks later over something else
So everything that's happening now everything that's being built. It's incredibly important
Um, which is why you know the chip process on bitcoin cash is is you know, the the greatest consensus model

(26:52):
We've got right now and it is you know, there's a lot of this work, you know
And I know you have highly respected within the community have a lot of regard
Um, and and nevertheless, even though you have already proved
Proved yourself in the ecosystem when a new chip comes out, of course people get excited, especially non-technical people like myself and i'm not a developer

(27:13):
Um, but nevertheless those chips are taken and they're ripped apart and they're you know, they're analyzed they're criticized and and rightfully so
Because this is important. Um, and and we want to make sure that we get this right
And i'm just thinking about cash tokens. Of course, it's a great great addition to the bch ecosystem
But I remember, you know the very first time I read about this was uh pm, uh, uh prediction markets version three

(27:38):
Uh, which of course then if you think about the same thing happened ripped apart
What came out of that was uh, of course much better and so jason's like just you know, because you're talking about opiaval
Um, and yeah be really good to focus in on this one and zonen this one just thinking about the consensus
Um, so the feedback that i've heard expressed by some bch developers

(28:04):
Um is that from their perspective the benefits are limited, but it brings a lot of risk
Um, and you know, i'm just looking at so this the opiaval there's a great, uh,
Uh mention of it from mike herne on bitcoin talk.org
Uh in october the 7th 2011
Um, and uh, you know, he also expresses risk here and i'm just going to give this quotation. I would love to hear

(28:28):
You know like what your how you would count that so this is a quote from mike herne
2011 the current bitcoin design has been carefully reviewed and found to be strong
People like kabinsky have tried the best and not discovered any issues
major structural design changes like opiaval invalidate all
All those assurances and put bitcoin back at square one

(28:52):
I'm not seeing seeing the benefit this this base is cheap and pruning would reclaim far more than shrinking transactions by a small amount
Lightweight clients don't even care. So this is of course, this is a long time ago
um, and um, and the world's moved on from there, but there's pretty like just taking that
Statement as it is. It's a pretty like damning assessment of opiaval

(29:15):
Well, what would your chat? What how would you how would you answer that if that was being said by mike today?
Yeah, um, well in first as you're saying there, um, but
you know, uh
He he did not have the information we have now and he was definitely right in in 20
I think that was 2011. Maybe it was 2012. Yeah, yeah in 2011 2011. Um,

(29:40):
And uh, he was definitely right in 2011. Um,
um, and I should even go back and say that uh,
Satoshi made a a
huge number of just incredibly prescient
decisions like um
it
I I am I i'm continuously, uh

(30:02):
Surprised at how good of a choice it was to base our virtual machine on on fourth
um
like
It may be because you know fourth is not something that sort of my generation of programmers
Grew up on a little bit and so I feel like i'm you know
in a lot of ways discovering lost ancient knowledge, um, but uh

(30:25):
He he really did pick the right tool for the job in expressing virtual machine contracts
Uh, because we can actually get quite a bit more concise code if it's expressed in fourth
Then if we for example implemented something, you know, if we had web assembly was was our virtual machine system
um
It would be significantly less concise and for no performance benefit and people do not under people have not thought through

(30:49):
These problems they just assume they're like, oh web assembly's fast throw it in there. That's the right thing to use
um, but like
You are throwing away 50 years of computing knowledge
um
So there I want to be really clear that I I would not judge any of the uh, any of the opinions
From you know from before

(31:11):
2013 um
By any of our of our standards now
We know so much more about what could be useful in the pro and in our programming model
Like what kinds of applications people are going to want to build?
Um, they're not if you if you read through that that bitcoin talk forum post
Because they're not even thinking about you know things like loan protocols and I mean

(31:31):
There's some sense of like hey we could eventually tokenize all the things and then decentralized exchanges
But that's like where it ends there. There is there's certainly no one there
uh considering that it will be practical at some point in the future to
Implement zero knowledge proofs using the contract system
Like there's just a foregone conclusion that it will be an opcode and I too

(31:53):
Bought that foregone conclusion until like
Within the past two years, um for sure, uh, the
even
even last year I was I was
Surprised to see to notice how uh, how practical it could be to implement
um some you know

(32:14):
Something like zcash's halo 2
um
Uh zero knowledge proof construction stuff in bitcoin cash as a covenant
um like that would have been if you had told me that in 2021, I would have just laughed at you but like
um
For sure, it's it's uh, it's viable. Um in the contract system and it can be expressed pretty concisely. Um

(32:38):
the
the
You you do lose like if we're talking about um about performance you you do lose some overhead on the on the
the actual um kind of manipulating bytes in in lots of ways but um
For for the long tail of of various use cases out there

(33:01):
It would be really really really hard to beat, you know satoshi's idea
That giving people a a virtual machine that they can that they can work with
That is that the system is the system is protected itself from people making silly programs
But people are able to just permissionlessly go off and build what they want to build
Um, we can actually get shockingly good performance out of it and you can build even crypto

(33:24):
applications
uh in this virtual machine system and and they're
pretty darn good and they they also scale really well because the you know, the
A lot of the biggest issue with scaling is not compute. It's bandwidth. Um
Um compute we've got uh, we've got a a ceiling we've got a a safety

(33:46):
Margin of 10 to 100 x right now
You know a a completely normal consumer device should be able to handle the worst case scenarios
Using less than 10 percent of their of their compute power
Validating the bitcoin cash network. So it's then that's that's very

(34:06):
Very normal consumer hardware and it gets better every year. So we're in a really good spot there. Um
Our problem is not compute speed it is simply bandwidth it is simply how many bytes it takes to express things
So, uh, that's that's what if anything when i've and i've kind of harped on um, all right, i've given i've given

(34:31):
privacy
Privacy applications as an example for some of this stuff because it's something really tangible that people can can understand quickly
And and I think hopefully it helps to kind of break the the misconceptions
around
What what is reasonable to do in the vm?
bytecode, um, it is reasonable to do crypto in the vm bytecode and not in like the you know,

(34:56):
the way that uh
Earlier iterations of sort of the bsva ecosystem
Would talk about like oh we can do signature checking in in the script
You just have to add these three megabytes of code or whatever it was
We can do it really efficiently they can be small atomic transactions that don't look much different than a multi-signature transaction
like people people need to really understand that like the

(35:19):
There are there are
relatively straightforward
Optimizations that can be made to get our contract system to a place where you can implement anything that is being done
By any layer one coin out there you can do it on bitcoin cash as a covenant
And you can do it with the same efficiency that is being done by those layer one networks

(35:41):
Like if you specifically designed a transaction format to be as efficient as possible to do a zero knowledge shielded transaction
The the total number of bytes taken up by that transaction will be very similar to to the optimized version
Of a bitcoin cash covenant
Yeah, and for me this that that all sounds for me. That sounds incredibly important when we're at that sort of

(36:03):
um, uh
bandwidth
Layer, so if you know, we're really pushing forward, you know, maybe we're heading towards our first hundred million or a billion
um, what what interests me today is like just thinking about um, like the realities of
Of bitcoin cash today where we have um now with abla we have this adjustable block size

(36:24):
Um, but we're still you know, the average block size
I don't know what it is if it's 200k or 100k
But it's basically so currently the the throughput is not the bottleneck today, you know, of course
I know you're looking ahead in the future
um, but just as a real realism today
What what benefits so basically so at the moment, of course all that transaction size is is also um,

(36:48):
Uh a subsidy paid through for the miners for actually using up that block space
So if we made that efficiency more efficient today with such low transaction volume
um
Could it not also so so what I also see is the impact of that is that the block size is is smaller now
Of course, it's more efficient. That's all great when you've got the transactions, but in today's reality
Um, I can see that there's sort of some of the negative sides to that but the the benefits then come

(37:14):
Later on it so my question would be why why implement this now?
As opposed to you know
Later on when it's actually looking like okay, this is on the horizon now that this efficiency and bandwidth is required. Yeah, um,
Yes, like if I if I understand the question correctly
Why why make transactions more efficient when our blocks aren't full and we want to get more data in the blocks to pay?

(37:37):
To today today is the question. Yes. Yeah, so of course I see the bad benefits of that. Yeah
And and that is just not today. Yeah
Um, and that is the argument that is driving btc today
The argument that drives btc today is that we want to um, you know maximize the transaction fees
uh in the immediate term with with uh

(38:00):
uh ignoring the availability of alternatives and and substitute products and um
and so as many people in bitcoin cash have discovered obviously that um
There are use cases that are that are prevented if you require
Large transaction fees for the use case like you're going to choose to use a different system

(38:25):
other than bitcoin cash if you're if you're
Uh per transaction cost is non-negligible basically if your per transaction cost is like more than a dollar
You're thinking about ways you can switch to another system that doesn't cost you so much
Um, and and this was this was just like the the critical and obvious flaw in the whole

(38:46):
we're going to make every transaction cost a thousand dollars and and uh, and a
A transaction on bitcoin is going to be as rare and special as chartering an oil tanker like this
Absolutely delusional takes um
No, what's gonna happen is they're gonna switch to a spreadsheet all the people that have to pay those fees are gonna switch to a spreadsheet

(39:10):
Yeah, this is what we were joking about this is maybe not even a spreadsheet. It's just a piece of paper on michael saylors
And the same is exactly true of of any use case that can technically be done
on bitcoin cash covenants, but are
Uh absurd to do bitcoin cash covenants, right?
If it if it requires 20 megabytes of transactions to to run one zero knowledge proof

(39:36):
You're not actually going to do that like you might do it for demo purposes
Like I might do it just to just to prove to everybody i'm smart enough to write a zero conf
Zero proof in this system, but I can make it work. Look I code golfed it guys
Um, but like no, it doesn't actually matter. It doesn't move the needle for anybody on the network
Um, if if the per transaction cost is absurd

(39:59):
I've done nothing but but prove to you that I that I can do it
It doesn't it doesn't translate into real use
Real users aren't going to spend, you know 20 or 50 dollars or whatever it is per transaction
Instead we want to find ways
To make more and more use cases really reasonable on bitcoin cash

(40:21):
So that people actually want to do them makes a lot of sense to me
So yeah, but so but so is that what you're saying?
So that's without opi value that the things that would make possible there really is taking like several orders of magnitude of data
Down is that the case? Yes, uh insane. Yeah, both opi value because um, you
Uh, you have to understand there are two things that are kind of are are huge limiters right now

(40:45):
One is function application and the other is loops
um, and you know, technically we could say that it's actually just function application because you can always do recursive functions, right so
And and we we can recurse down to a hundred, uh a depth of 100 in our current vm limit system. No problem
It's no, uh, no issue at all already handled by the vm limits
Um, but loops are also like 25 more efficient for a vast majority of cases

(41:11):
So it would be kind of silly to have obi val without out without also adding loops
um
The inverse is also technically true. It's technically possible to refactor a program
um to to specify the program as a a giant
Set or series of loops

(41:31):
um
where you
Effectively activate and deactivate different sections of the code at just the right moment to get it to do the things you wanted to do
As if you had function application like I think it is technically possible to say that they're equivalent
But like that's ridiculous. You should just have function application, too

(41:51):
um there we wouldn't even be talking about uh
about what you can get away with in a programming language if
We weren't starting with a severely
uh reduced
Programming language
Um, so yes, it is useful to have both function application and loops

(42:12):
For different for different reasons in different cases
Um in each case one is is going to be much more optimal than the other
Um, anything you can express as a loop is something around 20 to 30 percent more efficient to do with the loop
um
and then
It would be hard to even do the it would be hard to even give you a a percentage for

(42:36):
The the reverse where if you tried to specify the program as a giant loop and then activate and deactivate conditionals in the right
You you have to know exactly how many code branches your problem has and how many times you're going to hit that part of the loop
And where you're going to put stuff and by the way, you also need a compiler that's smart enough to produce these programs
And I don't know who you plan on making that. Um

(42:57):
Um, but uh, it's it's hard to do just just by hand. Um, it it and and i'm not even i'm not even
Completely certain that it's possible to say it's always possible
It's totally conjecture, but I think it's also it's always possible
Um to to specify it that way. It's just ridiculous. It would be very silly to do it that way. Um

(43:19):
Yeah, so when when looking at like so
Uh out of the the free chips that you propose I believe in the yeah, no all in december. Sorry
Um loops is the one that that seems you know from what i've seen in in the developer chats
this one seems to have absolutely the the majority of the support that this is like a really something that can be really um

(43:41):
Beneficial to bch the uh, uh, the risk reward is very nicely balanced
One thing that interests me, um, and I didn't see an answer to this is the chip and unless i've written it down wrong
But the chip is two thousand twenty one zero five
Um, so it's it's dated three years before the other two
So yeah, is there a reason like why why was this left untouched for so long? Why has it just been picked up?

(44:05):
um and now and and and and uh proposed for the upgrade yeah, um, it was waiting on limits, uh,
All of these all of this stuff was waiting on a reasonable solution to limits our limits were this, you know
implicitly defined by magic constants and encoding sizes and now they're explicit and so um
That there it's possible for us to um, actually go all the way back to um to my current comment there

(44:30):
He said it very well in 2011 like we we are certain that the system is safe as it exists today
If you change anything about the behavior of the system
Um, you have essentially thrown away all of those guarantees
um
and the most applicable upgrade
to to to point at that to compare to that

(44:52):
um
To that statement is the vm limits upgrade the vm limits upgrade
uh by necessity it's not it's not possible to fix them without
Necessarily exposing yourself to throwing away all of the security guarantees of what we had before
Yeah, so to be very clear, um the the the mountain to get over was vm limits

(45:15):
Um, and it took three years, uh, and and and that's just three years that the chip after the chip was proposed
Um, and in practice it took me all of last year literally all of last year and several other people reviewing very carefully many times
Um, uh, and that's just the the chip

(45:38):
Before then we have another decade of people looking at this problem, right from 2011 to 2021
There's a big gap. There's a whole decade gap between
Um, when you know when mike hern says that you know any change to this
Necessarily throws away our security guarantees. Um, that's where it necessarily throws away our safety guarantees rather

(45:58):
Um
uh
So you have to you know, there's
We had a decade a decade of thinking about it if you will before that chip was proposed and then the chip also in 2021
um
uh
Was you know
Maybe 50 of the way there but all of the work was in the last 20 percent

(46:22):
Um, and and that was a lot of last year's work. So um, yeah the
the
the the huge risky thing
Always is going to be
Modifying how the virtual machine applies limits
um that that is the
uh

(46:42):
That is the underlying concern in my current comment there
It was the underlying concern in why we you know, anytime anyways, like why can't we just have loops like
It would it doesn't change anything if I do if I do a loop versus copying pasting the code 10 times
Like can't we just do loops that let you loop 10 times and that'd be fine
and but the answer was was always well, yes, but um

(47:04):
actually our limits are all implicitly defined and based on based on magic constants and and uh transaction encoding, so
um, even if you did it that way you get these unpredictable spikes in the worst case in
in
Places that are that are hard to even reason about
Um, and so in practice the answer is just like no we probably can't do that. Not unless we do something more reasonable for our limits

(47:28):
um
and so a lot of contract designers have
Obviously, it's like the first thing you're going to encounter if you're doing anything non-trivial
Is that you can't like do the same thing three times?
Like you can't check the number of inputs and do the same thing across those three inputs
um
like
Uh, so a lot of contract authors

(47:50):
um
Have you know direct real world experience where they have already had issues with um with loop
Uh with not having loops
um, i'd say this the op eval, um
I think a lot of contract authors are a little bit insulated from the issue because they're using they're using cash script

(48:14):
And and cash script lets them not really think too much about what's coming out the other end
Of the system so, you know
It's easy for them to not really notice that
Hey, my my my compiled transaction
Has 80 percent?
Uh, 80 percent of the code in my compiled transaction is just copy pasted stuff

(48:37):
um
And and not just for loops it's for if you do need to do a
A certain kind of transformation in two different spots or two different part code paths
Um that that don't necessarily take place one after the other that you can't you can't rearrange how you use the stack
To to turn them into a loop
um

(48:58):
It's just you know, fun programs use functions
so, um
Yeah, uh, I would say that's that's why you see
Almost anyone who's worked at all with a version with with contracts is
Is is going to notice the first thing they have to learn about
about bitcoin caches virtual machine

(49:18):
Uh system is that it doesn't have loops because they need them and they don't get them so they have to work around that
Yeah, sure or in some cases they just don't
Yeah, and and you say who who would no one would program in that if it wasn't this amazing thing of money
Uh, so it's you know, this isn't it wasn't created because it was a great design choice from a from a coding perspective
And I appreciate that and it's it's also interesting. So I was aware

(49:41):
um that
um that the uh
The upgrade this year that allowed then further chips to come on but it's it's yeah that now makes sense that that's why
Loops was parked um up until now. It's actually a great, uh, great time
I don't know if you're aware of this
So we're actually using cash tokens on phoenix and friends of technology, uh that you made possible with your chip

(50:03):
um, and uh for um, there's a token called the
Um emerald token, which you can actually use for um having a fiendish shout out
And I think this fiendish shout out is that is well timed
Uh, just thinking about that particular upgrade which is coming this may
Oh, i've lost it now. There you are

(50:26):
builder entrepreneur or freedom cash fan tickets for bch bliss are on sale now check out
www.bliss.cash forward slash 2025 you can get the tickets on tap swap
dot cash
um, yeah, so this is for everyone that's listening bch bliss conference in may we're going to be uh,

(50:47):
so you've got a lot of builders coming there and this is planned actually around the
Uh virtual machine limit upgrades and big int so the chips are going on to main net
um, and and allows then the the chips that jason is talking about
Uh to be possible, you know to even to even be able to be uh implemented onto bitcoin cash if it passes through the consensus model

(51:09):
Uh, and there's a general consensus that these chips uh add enough value
um to the ecosystem
Uh to warrant actually putting it in so jason just sorry just to sort of branch off a little bit just quickly on that note
um, you know, these are your chips that are being
Upgraded the the conference is a celebration of that and while also showcasing the the building that's going on in the ecosystem

(51:33):
Peter ricin is also there joining to talk about his hardware and scaling of utxo
Uh, is there any chance we'll be seeing yourself at bliss this may?
No, I hope I can make it I I might
Uh might not be able to but I hope I can I can make it somehow. Uh, maybe digital
Okay, digital would would be okay

(51:53):
We can accept this if you do if you do turn up, I I promise you at least one beer is on me
So that's a pretty big offer
Uh, it'd be great because yeah, we I don't think we've no we definitely haven't met uh met in person, uh, but it'd be lovely
You know, uh, if you could join so I was there, uh, uh last year

(52:16):
And I think like all the other builders that were there. It was a a tremendous experience
So if something happened and you can make it to come then you'll be of course as you already know very very warmly welcome
Well, thank you very much. Yeah, i'm i'm glad glad to see uh, it glad to see it happening. It's it's exciting to uh,
To see the uh conferences beginning again sort of yes

(52:39):
Yeah, definitely. It's important. You know, we can uh, uh
Uh sit at home and we can code and and you know try and make the world world better from from our
bedrooms or from our computer but
We have you know, the really important thing of bitcoin cash is also getting out there meeting each other
And and forming enterprises is what I always find interesting is you've done a lot of work

(53:02):
um for helping the
The chain and developing the chain and making it so that more things are possible
on the chain
and uh
I think with these sort of conferences is a chance for actually people to come together the developers but also then the entrepreneurs
In okay, how do we leverage this?
To you know to make our business to make something profitable to make something that we can then grow the network even more

(53:28):
And this is the thing that uh that excites me. Um, so uh, yeah, but I I guess this is the
The uh working together with entrepreneurs to build products and apps. I guess this isn't your thing jason
Uh, i've yeah, i've been focused on uh
on protocol improvements that I I consider pretty foundational but um, i'm actually very excited i'm i'm certainly a little burned out on uh,

(53:55):
on protocol things and i i'm excited to be spending more of my year on uh, on
on software that I can ship myself and and don't have to uh,
Don't have to advocate for so much
Yeah, so yeah, i can imagine bit off id live off chain graph are all getting a lot more attention from me

(54:15):
Uh already this month and the rest of this year for sure. Okay, um, and I should say let me if I can also mention also that like
um
In some ways, that's why the txv5 chip, uh looks the way it does. Um,
The txv5 chip includes a lot of things. Um
That are in some cases, you know, have been my notes for like the better part of a decade. Um, and

(54:42):
Uh, I wanted to make sure that I got them specified somewhere. Um,
where we can kind of serve a little bit as a as a checkpoint for um,
uh
If if we could if we could um
Deploy a new transaction format. This is what I think it would look like
this year, um and

(55:05):
So i've posted in a few spots. Um, I answered some questions on on
twitter yesterday, but uh, I I don't
See the txv5 chip
Being a good idea to deploy
In 2027 unless bitcoin cash has has made it back into at least the top 10 cryptocurrencies. Um,

(55:26):
because uh
And the way I put it yesterday and I put it before is that um, you know transaction format version
Is very very hard to do because it's a natural bottleneck for for um a lot more ecosystem players like these 2026 chips
Essentially just uh, they require only node upgrade
Only node upgrades. So, uh, they're the kind of the least

(55:51):
Uh the the the lowest cost and and lowest risk kind of upgrade
Um, and they then they have uh, and so their risk return ratio is a lot higher whereas a transaction format upgrade is really
Really disruptive. Um, I have done a lot of review before
On specifically what software would be would be broken by the pv3 upgrade actually in 2021

(56:16):
and I was
A little surprised actually at how few wallets actually consume the transaction format
There are very few wallets out there that are that are decoding transactions
Most of them are asking their asking their back-end servers for utxos directly and they're coming back in json format

(56:36):
Or something like that. And so in practice what we might assume would be breaking with the transaction format is not quite as breaking as
as uh
As we might otherwise assume but in general transaction format upgrades are really really really disruptive
And they're a natural place where any any company that's got like just uh, kind of has a bitcoin cash integration

(57:00):
In their in their you know block explorer software
or something like that and they they it's so it works with bitcoin cash right now, but if the transaction format was changed like they
would just just as
Likely remove bitcoin cash support rather than doing the even if it's just a little bit of work
rather than doing the work to um

(57:20):
You know fix the the new part of whatever transaction format is changed. So, um, that is that is just a problem of being
Being relatively small. Um, maybe
Being being relatively small. Um, no one is dropping ethereum over ethereum upgrades
You know, they might be dropping ethereum, but it's not it's certainly not because of a transaction

(57:43):
some some uh minor improvement that or minor change in how a transaction is encoded, um,
You know dropping ethereum requires
If they're they're making a different strategy for how their business operates, right?
Um, and and my hope is that at some point, um bitcoin cash will be there
That bitcoin cash will be in a position to deploy a transaction format upgrade

(58:07):
Um, and if that were the case, I hope that this chip will have uh will have uh aged like wine
To that point
It's basically it sounds like this is a little bit of a present uh for the community while you sort of take a break and
and hoping you know, just just during this uh,
This episode bitcoin cash is written actually like seven seven to ten bucks

(58:32):
Uh, so maybe we'll be back in the top ten by the end by the time we finish right for carrying like that
um, but yeah, but it but you you've got something in your mind that this is really only when um,
The the ecosystem is willing to swallow
The the effort and pain of making a transaction a change to the transactions which makes sense
What what interests me was um, like the

(58:54):
Some of the other chips they seemed much more specific and and smaller
Um, maybe not in the scope of what I actually enabled but but actually you know in the format
Whereas this chip you've got multiple different aspects. You've got the zero overhead commonance detached signatures fractional satashis
Um, and there was you know, I heard some groans from from developers at jesus, you know, this is an omnibus

(59:17):
the bible of chips and and um,
um, uh
And just raising an eyebrow at that. So, you know from your perspective
Like why why have all these bundled together?
Is it that they're linked is it is it just for you for the amount of effort or is there a reason why they couldn't?
Have just just been
broken up into separate chips
um

(59:38):
Uh, yeah, and uh, so some of that is definitely justified there are
This could be three chips. Um, there are two two
other components that are
Uh, very reasonably broken into a separate chip
um, i'm
Not interested in proposing them right now. Um, the 2026 chips are definitely the 80% of the value for 20% of the work

(01:00:02):
So the the chips that I proposed for 2026
uh, the the ones that I proposed in december are the ones that I think um
Get us a huge amount of value for uh for very safe very contained and
Demonstrably will stand the test of time
uh

(01:00:22):
changes
the
2027
The stuff in the in that 2027 proposal the the transaction format proposal are things that um
uh
the uh
most of the transaction format changes
really truly just

(01:00:42):
improve the compression of transactions a little bit like there there are 20 improvement versus
op-eval and loops are a
1000 or 10 000 x improvement
and so
It's not even reasonable to compare them and how and how
Irrelevant the the differences are in in some of the transaction encoding stuff versus op-eval or or loops

(01:01:06):
um, sure and so I
Hear that and this is also kind of interesting for like for me. I'm more of a like from the pragmatic side
um
Is so what what I hear is um
To implement this it requires really like some swallowing
Of from from all the other stakeholders because they have to you know, it breaks basically everything
Yeah for for very for very little I have to say if I wanted to go out and selling this I wouldn't take you

(01:01:34):
Yeah, it's um, it's something that we should we should have
It's important, but it but it's definitely not urgent
there are a couple with a couple exceptions inside there that there's a
Read-only inputs are are a backwards could be implemented as a very
Isolated tight backwards compatible thing. They're definitely not as as important as the stuff I propose for 2026

(01:01:59):
That's why it's in the 2027 stuff and it's the 2027 stuff is sort of like and this is stuff
uh, that's on the horizon that I i'd like to start the conversation for um, but but um, and
I definitely don't consider to be uh important enough to draw attention away from the stuff that I propose for 2026 like

(01:02:19):
Uh, if there's if there's anything that's worth spending spending your your
Your brain power working on it's the it's the thousand x and 10 000 x improvements you get out of the 2026 stuff
um
The read-only read-only inputs are so I I would say you could break the 2027 stuff
um, you could break it into three chips that the

(01:02:41):
The read-only inputs can be a very isolated thing and they're a big chunk of the value in that
uh in in that chip also like they're
um
They would allow us to get to the zero overhead covenants. Um
I
I I wouldn't necessarily even be against trying to make that happen in 2026, but I I consider it to be a lot less

(01:03:05):
certain than
loops and eval and p2s simply because
Like loops and eval p2s. There really is just not a lot of space for bike shedding like there's there's um, there's a
Correct way to implement loops in the fourth programming language and it's been around for 50 years
Like you can ask ask a large language model to like as I mentioned earlier to to implement your program using

(01:03:28):
Op begin op until and it will do it right like um
It would it would be really hard to justify
Uh inventing our own thing for for loops
um and in the same way op eval
Does the simplest thing as we said before it says the simplest thing possible that we can prove
will never

(01:03:49):
be
uh
Will never become outdated
by a new uh
a new clever way of of approaching the op eval problem like
It will always be and we can prove it will always be the most efficient way to
Take one item from the stack and evaluate it to you

(01:04:10):
It's always the most efficient way to do a single function call or the last function the last call of a particular
function
So it's one of those it's it's the the one approach that
We can be certain would stand the test of time versus something that we tried to kind of invent ourselves if you will
Um, and then the p2s stuff similarly is quite literally unifying

(01:04:34):
Three limits so that they they are logically consistent like they're right now. They're not logically consistent
and so uh
of all of those
To me, uh feel like much more obvious things
Than anything that I kind of put in the txv5 stuff
um, and the txv5 stuff is uh, quite honestly in a lot of ways is like

(01:04:55):
I am I am I don't have it in me to break those out and like try to champion them and start the process like
They uh, I think I think everything in there are the correct ideas and they're the things we should probably be talking about for for the
the problems they they they address um
Read-only inputs could be its own thing and then the the bytecode limits could could be their own thing

(01:05:16):
Everything else actually does have to be its own chip. It does have to be one
transaction format upgrade
Um, and and that's that that is just the nature of a transaction format upgrade like it is necessarily an ominous an omnibus bill
Uh, you you can't just kind of break the transaction format
If you're gonna if you're gonna trim all of the dead weight from the transaction format, you should get it all right the first time

(01:05:40):
Yeah, sure. Um, and that's what that's what the rest of the chip is like for me again
This is just looking at the pragmatism of of bitcoin cash today. So, you know the the transaction, um,
Unchained transactions are relatively stable now
Since the end of the the bear market
So 2023 of course is that like there's some spikes of particular use cases that you know testing of new software

(01:06:05):
But the base layer baseline is pretty pretty stable
It's pretty pretty standard. So, um, you know, basically right now i'd say bitcoin cash the
Where I can imagine this would be useful is is for example if we do end up with the zero knowledge proof
um
So you can have as you put it's nicely like the monero like or you know, potentially even better you can have some

(01:06:28):
um
modern security
um privacy
layers on on on bitcoin cash
um, and maybe maybe this would generate a lot of transactions theoretically and you know,
for example if that was to happen and and bch was to cannibalize the entire dark market usage and and monero for example users would

(01:06:49):
Would use bitcoin cash missing. Okay, the
there's a real like this is being used and um, then what i'm the one I understand, um this chip is
Okay, if we were to implement this
We get to basically we get to say I don't I I don't know but I would you know
Whether it's 20 or 30 on on that transaction size

(01:07:10):
Uh, so that we can end up having more throughput
Um and still you know stay on the course to being a a world used
um currency even with those particular
Covenant's being used is yes
Yeah, um, and I'd go further a little bit a little bit of my intention of like just kind of getting it all in one
technical technical spec all the things we need to

(01:07:32):
Uh clean up to be able to put a literally put a bitcoin cash
Zero knowledge proof covenant transaction side by side with a z cash shielded transaction and compare them in byte length and and demonstrate
That the bitcoin cash transaction is the same length or shorter than the z cash transaction
um, and my my intention is to

(01:07:52):
Is to prove beyond the shadow of a doubt that that's possible to to just dispel the idea that you need
Layer one consensus upgrades to add privacy you don't um
I I wanted to get that in a technical spec so that I could I could point to exactly what changes we would need to make
To shave off every single bite
um

(01:08:13):
Uh required to to put those next to each other and say yes, we can do z cash on bitcoin cash
Uh at at no bandwidth cost
um
Uh at no no additional bandwidth cost and actually um
As I I mentioned in a few places in the chip too
There there are real advantages to doing it at a user layer where users can deploy

(01:08:38):
Their own protocols they can deploy covenants that implement a new protocol because there are there are you know
Bleeding edge ideas and a lot of these um zero knowledge proof
Uh fields where a a layer one network simply cannot deploy something
That is a white paper that came out three weeks ago

(01:08:59):
Bitcoin cash would be able to deploy it and people would be able to use it with 10 bucks and it would be fine if it broke
It wouldn't it wouldn't jeopardize the network and it would bring the it would bring
Um like university teams that are working on these things or or or new companies that that have
You know, they're they're coming from from the math side of the academic world

(01:09:24):
Um, whether they're coming from the zero knowledge proof academic world
And they're wanting to turn their turn their ideas into a product. Um
that uh, that is
Usable on a real cryptocurrency. I think bitcoin cash could be a
An ideal target for them to to
Essentially publish their white paper with a proof of uh with a proof in um, I I think that there that

(01:09:51):
There's there's certainly a future possible in the next few years where
Novels your knowledge proof constructions are our first
Demonstrated on bitcoin cash
And because because our vm will be capable of it and and it's actually unique to bitcoin cash because ethereum
It would be completely unreasonable to try to implement these things in ethereum because the transaction fees because ethereum has this kind of 1000x or worse

(01:10:16):
Scaling disadvantage versus bitcoin cash and there's no way to fix that. That's an architectural problem
To fix that you need to make it look like bitcoin cash
So which I think is which I think is such an important point, right?
So and that's why you know, I love what you're doing. I think it's important with the vision
It's why i'm with bitcoin cash is because
Technically, um, it's built for the future. Technically it's built

(01:10:41):
for
um for mass usage
Um, and also, you know, but but it's also on the flip side of that the build it and they will come thing
Um, it's clearly not true, right because ethereum
They've got this something, you know, that's not as scalable
But but it's being massively used bitcoin cash has this amazing ecosystem. That's massively more scalable

(01:11:01):
And it's used, you know a fraction of that which is where you know for me
It's important to get that balance of basically where where the resources are put where the effort is put in the ecosystem
To make sure that the users are are growing but before I'd like you know, we've got a few more things
I'd love to talk about i've got a final fina show out here
From bruno get a bunch of self-licensed cypher punks working on your privacy aria vpn underscore net

(01:11:27):
It's a strict no logs policy zero knowledge architecture vpn provider
Not in any of the 14 i countries bch zero conf and xmr only 30 bucks for a whole year plan
Business solutions available. So go and grab your vpn
With uh paid with bch or monero
So, um, yeah, this is jason then the intention of this by the way is to get transactions going

(01:11:50):
Because people can actually pay with uh with tokens on bitcoin cash
And they send it and it just basically just goes around in a flow, uh between uh between essentially my wallet and uh
And cauldron decks is to encourage people to actually use the tokens and use bitcoin cash. Um,
So, um, so this is a

(01:12:12):
Bitcoin cash. Um
So, I mean have you got any ideas on that yourself? Like where do you see?
Um the big growth coming from in bitcoin cash over the next, you know in the in the short to midterm
Um, I would say that the the all of the important work to be done there is actually on the entrepreneurs

(01:12:37):
Uh and and you know specific products side
I've spent a lot of time on these protocol upgrades, uh, because they're they're very foundational and they they unlock a lot of uh,
A lot of possible entrepreneurship. Um, but yes, i'm i am personally very excited to not be working on protocol upgrades

(01:13:00):
So i'm uh, i'm quite done. Um
and uh as as I
Like, you know kicking out the txv5 stuff, um, probably indicated. Um
I uh, yeah, I will personally be working on um
on chain graph and and libauth are

(01:13:21):
Tooling to make um
Kind of advanced wallet development easier. Um and and more advanced, uh
decentralized applications
Decentralized applications that you can use in across multiple wallets
Um, I think that's a big a big field that bitcoin cash is uniquely well positioned to to to kind of conquer

(01:13:41):
Whereas in ethereum and these other ecosystems like most most of the world is kind of based around everybody just
Builds their app and a website and pretends it's decentralized
um
so I I think that there's um, I think there's some
some
Some unique advantages in in how bitcoin cash works
Um that are just going to require a lot of a lot of more development and work

(01:14:05):
to
uh to to make them uh really
uh make them really into into practical products that that people can use but um
I won't I won't pretend to be uh good at
Identifying which specific entrepreneurial, uh ideas are going to are going to rise in the marketplace and and who's going to make it happen?

(01:14:27):
Because I mean ultimately in practice, uh, it's the it's
The individuals that are that are building these companies and products that are going to make all the difference. Um
and so uh
The the goal of any of this protocol upgrade stuff is is basically just to make sure that we've fully gotten out of their way
that the silly things that that are kind of just

(01:14:50):
you know
historical baggage in our in the protocol, um where
you know with in the case of limits we we kind of um,
You know satoshi basically kind of patch stuff up real quick duct tape some stuff and then left
um and uh and you know, we kind of we've been we've dealt with that for quite some time
Um getting that kind of in a in a better spot where people can use contracts to kind of the extent they expect

(01:15:16):
The the thing you know, everyone is in the community is quite well aware of how
BTC is
Is look down on bitcoin cash and like to you know
Joke about every aspect of it, but that's changed quite a lot recently when it comes to the development and tooling that's
Happening on bitcoin cash. I guess you've seen this on on twitter

(01:15:39):
um, and uh
Yeah, this is definitely an interesting change that people people are becoming more aware
Um in the ecosystem in the wider ecosystem outside of bitcoin cash. Most importantly
That okay things are cooking here. Like these guys are really improving
This isn't just bigger blocks because that's that's what most people think right bitcoin cash is just bitcoin with bigger blocks

(01:16:00):
Well, not anymore
Um, yeah, there's uh
Yeah, i'm excited to especially this next couple years are going to be are going to be exciting some of this stuff
um, you know as we've seen with cash token it cash tokens it
Took more than a year to get um
um, some of the some of the software in place even though we had you know, we had a

(01:16:23):
patches for a block explorer and patches for a wallet or two
um before
Activation, um, it's just really really hard to
Speculatively work on products that were that rely on a consensus upgrade
Um until it's until it's locked in at least locked in and and in practice a lot of companies are still going to be

(01:16:45):
uncertain until after may 15th
um, because that's what they've come to expect for from
the the wider
Specifically the wider bitcoin ecosystem, but also just all cryptocurrencies like you don't know it
You don't know it's going to be it's going to be a stable ground until it's for sure there
Um, and you can't you can't waste your company's resources

(01:17:08):
Building building for something that that might not ever work
um
so having having
The kind of protocol upgrades
actually live and working, um
Is is a necessary prerequisite for the the kinds of efforts that are you know, uh a year-long technical development effort

(01:17:30):
producing a product that that is
That that could really it
Uh could really improve our bottom line on that on that layer one usage that we were talking about
um, so and what what are the ideas
Was that sorry you said I mean you say you're not an entrepreneur
But you've come up with some some nice ideas that could at the very least be used by entrepreneurs

(01:17:50):
And one of the you know, it was that prediction markets, you know, which we're seeing uh, to some degree with bch guru
although, you know the prediction markets, you know, I as understood you had had in mind and I think that really interests me is
Um is wider than that so, you know to actually start sort of
validating truth or what is the truth or putting some money behind actually what the truth is and

(01:18:11):
um, which is very interesting but but one of the things you had was jadex
Which you know, you saw a lot of opportunity here
uh for people to very easily, uh
Sell their tickets and set up basically a system where they can where they can uh distribute and uh
Set up their own exchange for for just pennies

(01:18:33):
Um, is that something that's been parked?
Is it is it shelved now or is it something that is if potentially, you know, you would continue working on at some point?
um, yeah, certainly, uh, it's uh
No, um, everything was everything was parked
except for vm limits
Uh for essentially all of 2024. Uh, so as far as my personal my personal development schedule

(01:18:56):
Including including basically nothing other than vm limits. Uh, in fact my real life schedule included almost nothing but vm limits last year, but um
Um, but uh, by the way, which thank you, you know, really it's very highly appreciated
Work in the in the ecosystem. Thanks. Um
But uh, yeah, so 2025 i'm definitely excited to be working on um on

(01:19:20):
more, uh
essentially just
uh
direct products, um, and yeah, I I have a lot of work to do in chain graph and I've uh,
A lot of work to do in lib off
um the javascript library, um
To uh make some of these things that are until now have been kind of demos

(01:19:41):
To make them really practical and easy to implement in wallets and and uh easy to kind of copy templates between wallets and just run it
Um, yeah, there's there's um, there's some very exciting stuff to be done there. Um, and i'll be working on that
And uh
At some point when that stuff's good enough then then actually the jet x template from 2022 will just work

(01:20:05):
so, uh in any wallet that implements, uh the wallet template stuff, so um
The the design of that is
Uh, you know 80 of the work done and with big end. Um, the last 20 is simply removed
Um, you you don't need it. So, uh, yeah, so the the 2022
Um implementation is actually already fully optimized and ready to go but getting from you know,

(01:20:32):
A theoretical protocol spec like that to something where you click a button in a wallet and it actually works
Um that that's a lot of work and and it's not just like tooling work
But also there's some entrepreneurship there of like, uh getting getting
Wallets in in the hands of users that are wanting to use it in that way
Um, so yeah, I I definitely expect to work on um the

(01:20:55):
the
um wallet engine in libauth over the next um
You know into the future
indefinitely, um now uh and and uh when that is um
Sufficiently capable it will be possible for it to just execute the jet x template from 2022. Um, so yeah

(01:21:17):
Anyways, if jettix specifically is is as done as it can get
Until you get to the to the wallet integration side and and then there's a lot of wallet integration tooling work to be done
And i'm i'm working on that now. So um, yeah, we'll see
That's what we want to hear I mean this is this is the thing that I I think um

(01:21:38):
It is clear in bitcoin gash is the entrepreneur side. This is that the people that have got to you know, the eye of the resources
to um
to use what's what's there?
um all of the the possibilities on the chain all of the work or the groundwork that you've set out already on the
um on the main chain to to then extract value, uh rather well offer value right um from from the chain, um, and

(01:22:05):
Because you know, there's always the the things coming up, you know how to get bitcoin cash. Um
Used more we need more marketing and so well, you know, my question is well who's going to pay for that?
Uh, whereas of course you have entrepreneurs selling their product or service
Um, then they're they're paying for the marketing and and they're financially incentivized to do that
I think the biggest bitcoin cash marketers that I see

(01:22:27):
Um are general protocols, uh, you know, but you you see the any hedge comics going out you see uh,
You know always something and because they're marking their marketing their product. Um, and and I think this is the greatest way
That will see bitcoin cash use grow
Uh is through that so i'd love to see more entrepreneurship in the ecosystem and what I think is really important

(01:22:48):
Um, because as a non-developer it's uh, and also I think it's just also someone that's that's not got a business that's running on
um, the bitcoin cash like directly
Is is to make sure that the chips that are being put in are actually based on economical need
um
you know because the

(01:23:10):
When you have entrepreneurs that are that are building things then of course then that feedback the the mix between entrepreneurs their needs for
For business that is really
Utilizing bch as money and everything that it offers
um
That's actually a direct economical factor of hey, you know
This has a reason why it should be adapted because by by adapting this is allows my business to

(01:23:32):
To be even bigger and grow faster and and use the chain even more than than it than it's being used today
um and and that's what i'd love to see just to make sure that this balance this balance is there because um,
You know, you're clearly an incredibly smart person and you're looking very far into the future
um
But nothing beats cold hard reality

(01:23:53):
What's what's what's actually happening today? You know how people are using things? I think in the sort of feedback mechanism
Um, which I guess we could do a bit more of today
Um, yeah, a lot of very smart people have been completely ruined by mr market so
Yeah
Yes, I definitely agree i'm excited to see lots of bitcoin cash companies, um,

(01:24:18):
uh doing
building more and more uh products that are sort of unique to bitcoin cash and especially like uh token
Token related products where you need both good good token support reasonable contracting capabilities
And pretty low transaction fees. Um, and that's actually a mix. You can't really get anywhere else

(01:24:39):
You can kind of get it a little bit on solana right now
um
but uh certainly not at the scale and um
But yeah, not at the scale that uh that bitcoin cash could offer it so um, yeah, scale and transaction fee, uh, yeah reliability
Like get transactions that don't get uh, don't get cancelled
um, yeah, and and I guess uh also

(01:25:01):
Decentralized but um, yeah anyhow, um
Yeah, so i'm very excited to see that stuff happening and uh, and hopefully to to contribute a little bit of a little bit more of it myself
From that side a bit. Yeah, I've spent a lot. Yeah, I've spent a lot of time on
On you know, basically making our programming language support loops and stuff

(01:25:22):
Yeah, and it's really and it's really you know, everyone
Everyone expected after getting the lock in in november
Um everyone expected and certainly I did that we won't be seeing much of jason now for a while
You know whether it's going to be five months or six months. I'm sure he's going to you know
He's going to just calm down a bit and uh, just a few weeks later. Whoa

(01:25:46):
Three new chips, uh, you know, no piece of the wicked
Um, so I think you deserve you definitely deserve a break. I can imagine that the many developers in the community that go through
These chips and really pick them apart
Uh, they also
Will be happy with a break. I think it's exciting
You know, I think it's really great to have uh more ideas than we need and to be able to pick the best ones and throw

(01:26:07):
Away, um, you know more I think that's great
um, but you know what I also you know, as we've said I think uh,
Um, the the stage right so you know one of the one of the chips and I know you've mentioned and you said hey
We have to be in the top 10 and so forth
But one of the parts of the the the new chip that you sort of uh, the announced, uh, was it just yesterday?

(01:26:27):
Um, it is gonna know today. Yeah
Yeah, I mean this talk being a good forcing function. We're like, I should probably get that out. All right
Yeah, yeah
Yeah, and it's great and I love the you know, um the the reaction it
It creates in the community. It sort of gives the
I think quite a high signal of how you are regarded and people are excited

(01:26:47):
You know about what what's possible and what can be done and how are we going to improve this chain?
Uh, which is why I think such an exciting chain to work at because people want it to see
We want it to be the best money that that can ever that will ever
Exist. Um, and one of the things that you know, fractional satoshis. I looked at this and I saw some people going
Yeah, yeah with this is something, you know
I have to admit I laughed and I thought wow, you're bullish

(01:27:11):
Fractional satoshis, you know, there's a hundred million satoshis
Satoshi is in a bitcoin our bitcoin cash
Um, so, you know, you're already planning for eight billion people using this and transacting everything on this. So I mean
Yeah, so it just I mean
You know, it's a problem we should have a good answer for when it rises

(01:27:32):
Um, it would be good for it to for for us to have argued about it a little bit here in 2025
Um, and you know there and again, there are also been other proposals
Uh in the past couple years and then there are some proposals from kind of 2013 era
And I just wanted to kind of get on paper
What I consider to be a pretty um a pretty

(01:27:55):
Uh straightforward solution, um that optimizes the things that matter to optimize
Um, and uh, and some rationale for it. So just uh that so that conversation can be can be brewing for some time
Um, but yeah, certainly fractional satoshis are not urgent. Uh, if anything fractional cash tokens

(01:28:15):
Are are much more urgent in that there are lots of use cases that there's a there is a plausible
A plausible concern that you wouldn't quite have enough precision
Uh with with what cash tokens offer right now because cash tokens essentially offer the same
Precision as satoshis
Um, and by design that was intentional to make sure that you know, whatever we do with fractional satoshis would do the same

(01:28:37):
We'd do the same thing with fractional cash fractional cash tokens, but um
uh
the um
Yeah, I can I can certainly see fractional cash tokens being becoming
Relevant even a little bit before the fractional satoshis. Um, but then there's also you know, if we if we had a huge
uh uptick and and increase in layer one, uh volume, um

(01:29:02):
There's a chance that fractional satoshis would be useful essentially just for specifying transaction fees
Um, that's kind of the primary use anyways, um in in the much longer term, but um
There are and the chip talks a little bit about um
You know today's largest, uh financial markets use tick sizes that aren't quite representable. Um

(01:29:23):
Um, if you if you don't have a little bit better division, um
So yeah, you can look at the rationale there, but i'm glad to be uh, sort of done with that
for a while
For a while, let's see
One or two months and be like oh shit i'm bored i'm coming back
Roll up the sleeves and get cracking. Yeah, i'll continue. Um, I want to continue like, uh,

(01:29:49):
Um clarifying rationale and stuff like that for for all of these
Um, but yeah to be super clear the the 2026 stuff is stuff that I um
I that I think
is
far
um, you know potentially far less controversial and far more straight far straight forward like things that
um

(01:30:10):
I have easy easy proofs that um, they will stand the test of time
versus the stuff that I I didn't
Um, you know try to propose for 2026 is the stuff that uh, I can't easily prove is going to be the optimal design or or
or has has designed decisions that that are kind of subjective or or um,

(01:30:32):
that uh
There's it's possible that we might uh
Come up come up later with a slightly better way of doing something
Versus the 2026 stuff, uh is essentially all sort of
In some ways sort of ancient tech but also just timeless things like the loops and and obi val are like

(01:30:54):
You're not going to do better than taking one thing from the stack and evaluating it. Um
uh versus
Doing something more clever
um
so, uh, yeah, I i'm i'm hopeful that uh that i'm interested in in the continued review of the the 2026 stuff and
I'm going to keep working on on tooling to make it very useful. Um in in end user wallets, um,

(01:31:20):
because that's uh, that's
Where I planned where I plan to spend my time for much of this year, but um
but uh
yeah, and I I
I
I i'm excited that there there's already a lot of
um
A lot of interest obviously in the loop stuff. I think the opi val, um

(01:31:40):
opi val, um
was certainly surprising, um
uh
because uh
again very few people are exposed to function application in their contract design because they're using they're using cash script and they just kind of
Ignore what comes out the other end in a lot of cases like you know
you put in the cash script and you and you get whatever it comes out and and you know if you if you

(01:32:03):
Disassemble it and and
Consider how you might make it more efficient. Um
Um, you would find that you know, there are some there's some obvious, uh places where a function application would cut cut 50% of your transaction size or something
But you know, is it most people are not exposed to that problem because it doesn't show up
At the at the cash script side where you need a construct to pretend that you have loops

(01:32:25):
Yeah, sure, and the same is true of the pos stuff, but if it's came with you so uh emergent. So john neary from
general protocols he would love to
Uh, yeah say something. So yeah john, you're you've got the mic
Hey, I just wanted to make sure that it's clear that uh, I completely disagree with fiendish's point that it's okay to take a break

(01:32:46):
No breaks
And if I know Jason probably his brain if there's something important to be done it's not gonna let him take a break
Listen I was playing the long game. Jesus
Jesus
Of course he's not allowed to take a break but we can't say that out loud man

(01:33:11):
Also actually constructive um, so about the topic uh, jason was saying
Uh that it's really important to develop. Uh, you know, it doesn't matter as we've been saying for for several years now at gp
It doesn't matter
Uh how much power your chain has if it's not accessible to developers and users?
And so, you know, that's the whole concept of the native wallet and as long as we've been working and thinking and conceptualizing about it

(01:33:35):
Jason's been conceptualizing about it even longer, right? Like he's had this code out for what is it like eight years or?
Like the cli wallet anyway
For a very long time because he also 2018. Yeah, I think it was 2018, right? Yeah
Yeah, it was the earliest stuff at the end. Maybe 20. Yeah, I have some videos of it. Yeah, so

(01:33:55):
Yeah, we we totally agree with that and I was also in the process of writing an email
Uh to invite jason to our internal, uh, we have a like a bi-weekly meeting
We call xo connect right because that's our attempt at the at the native wallet
concept and so i'd really like uh, i'm really hoping jason will join us at xo connect and uh

(01:34:15):
Have us hash out some of these uh issues where
You know, we we've got the concepts and then now it's time to like really uh make the rubber
Meet the road and actually put these things out in people's hands
And so there's a lot of little details to iron out and we'd love to have you on jason
So I hope you'll you'll join us sweet. Yeah, i'd be happy to have don't know how much I can commit to but I could for sure

(01:34:37):
It'd be fun to uh get together a bit. Yeah. Yeah. All right. Anyway, thanks for the time. I appreciate it
Thanks, you're more than welcome. Thank you jason so much for coming on so
Uh, yeah, I I promised not to not to take too much more of your time than than an hour and a half
Um, so I think that's a good point just to sort of

(01:34:59):
Uh wrap wrap things up. So um, yeah, thanks again jason for your time
It's been really humbling to have you on the show today, uh, and an absolute pleasure
Can't stress enough appreciate all your work that you've done in making decentralized sound money a reality
um, and uh, yeah and all the time
And uh, yeah and all of this, you know, it's an important mission and you know separating money from the state is something that is going to transform

(01:35:26):
Money more than anything that's happened ever before in human history
And I believe that we're on the right path to to reach that and that's through your help and work. So, thank you
Amen and thank you and i'll try not to let that other stuff go to my head either
You can let it go to your head and then so when you want to take your break you like shit
But i'm a god here. I can continue working ridiculous. You see that's the strategy

(01:35:50):
It's a good one. It's gonna work. I can feel it
All right
That's it the txv5 is the last of the of the chip stuff I I have in my backlog. That's uh,
I'm gonna go work on software now
Good. All right. We'll let you off. We'll let you off and we'll welcome you back when you manage to get back in so
for all the listeners, make sure follow jason on twitter at

(01:36:13):
bitjason so j
S o n to keep up with all of the bch developments see see if he can manage this see if he can go the
The rest of the year without uh chip proposals
I'm sure you can make a bet against this on bch guru because I think there's gonna a lot of people that'll take that bet
To be honest, let's see. Um
If you enjoyed the show and want to show your appreciation

(01:36:34):
Please feel free to drop some bch or a fiendish token into the wallet the address is in the comments at the start of the spaces
Um, the next episode of fiendish and friends is again at four o'clock central european time next friday
And it's featuring joshua petty aka elon moist developer behind now defunct decentralized social media app twitch
Let's talk about bitcoin scaling and the future of pdcash and cryptocurrencies

(01:36:58):
So great, uh again, thanks for being here and jason this was fiendish and friends
I wish you all a good morning. Good day and good night. Take care

All Episodes

#11 - Jason Dreyzehner: 2026 CHIPS OP_EVAL, P2S, Loops and TX5

Episode Transcript

Popular Podcasts

On Purpose with Jay Shetty

Dateline NBC

24/7 News: The Latest

.css-15opob5{left:0;position:absolute;top:0.8rem;} All Episodes

.css-14f5ked{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:2;overflow:hidden;}#11 - Jason Dreyzehner: 2026 CHIPS OP_EVAL, P2S, Loops and TX5