February 20, 2025 • 44 mins
What if gaming could be as accessible and universal as music on Spotify? Olivier Avaro, CEO of Blacknut, joins us to discuss his revolutionary vision for cloud gaming. Olivier shares his journey from the Commodore 64 era to leading the charge in cloud gaming, detailing how Blacknut aims to provide a seamless and affordable gaming experience to a global audience with a library surpassing 700 games. We explore how the company’s innovative hybrid cloud architecture is set to transform gaming by optimizing performance and reducing costs, all while tackling challenges like internet instability in emerging markets.
The conversation takes a technical turn as Olivier unveils Blacknut’s cutting-edge server architecture. Learn how their unique CPU, GPU, and VPU setup allows for unprecedented concurrency, enabling up to 200 players to enjoy high-quality gaming simultaneously. This innovation is crucial in maintaining affordability while ensuring minimal latency and top-notch performance, an advancement that sets Blacknut apart in the competitive landscape of cloud gaming. We also break down the role of 5G in enhancing gaming experiences in regions with previously unstable internet, providing a roadmap for how technology can leapfrog infrastructure limitations.
Finally, we explore the broader landscape of cloud gaming's potential to democratize the gaming experience. Olivier shares insights on expanding services into new regions like South Africa, highlighting the logistics and partnerships required to make this transition smooth. This episode promises a fresh perspective on the future of gaming, emphasizing the exciting possibilities that streaming technology holds for players worldwide. Whether you're a gaming enthusiast or a curious observer of technology trends, Olivier's insights offer a glimpse into the dynamic evolution of the gaming industry.
Stay tuned for more in-depth insights on video technology, trends, and practical applications. Subscribe to Voices of Video: Inside the Tech for exclusive, hands-on knowledge from the experts. For more resources, visit Voices of Video.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:07):
Voices of Video.
So we are at the top of thehour and looks like we should
get started.
Olivier, are you ready to talkabout cloud gaming?
(00:29):
Absolutely ready, excellent,excellent.
Well, welcome to those who arejoining us live.
This is the May edition ofVoices of Video, and, if you
haven't joined us before, voicesof Video and if you haven't
joined us before, voices ofVideo is a conversation, or,
(00:50):
some might say, a real dialogue,not a podcast, I guess, a video
cast.
We go live on LinkedIn and alsoa lot of other platforms, and
we are talking each month withinnovators in the video space,
and so this month, I am superexcited to have Olivier Avaro,
(01:12):
who is the CEO of a companycalled Black Nut, and we are
talking about cloud gaming.
I will let Olivier tell us allabout what his company does, but
welcome to Voices of Video,olivier.
Speaker 2 (01:28):
Thanks a lot, mark, for the nice introduction.
So my name is Olivier Varro.
I'm the CEO of BlackNerd, which, in short, is doing two games,
what Spotify did for music, sowe are distributing games from
the cloud large, large catalogof games more than 700 games so
(01:50):
far and this, you know, for asimple subscription fee, right?
Uh, I was long time a gamer.
I enjoyed it a lot, you know,when I was a teenager.
I enjoyed a lot with friends,with my family, later with my
kids, and I started Black Nut in2016 with the big ambition to
actually bring this joy ofgaming, this good emotion, all
(02:12):
the positive value of, you know,playing together, to the mass
market.
We deployed the tech for aboutthree years, I think cloud
gaming is, you know, doesrequire a bit of technology to
work efficiently.
Then we started deploying itall over the world, and this is
where we are today.
Speaker 1 (02:30):
I love it.
So I have to ask the question.
You know, sometimes when we'rebuilding advanced technologies
we get so into the technology wedon't get to do the thing that
we originally set out to do,like play games.
So are you still a gamer?
Well, I'm you set aside timeeach day to play or I set aside
(02:52):
each time to play a little bit.
Speaker 2 (02:53):
That's true.
And I have to say that, youknow I was the first game I
played was, you know, on theCommodore 64 machine.
It was then Bulldash, right, sothe older of the audience will
know about it.
Now I'm still.
I've been playing with my kid,you know, of course, on the Wii.
You know all the Nintendo gamesand Mario and Super Mario Kart
(03:18):
and Super Mario Galaxy, right,and to be truly honest, you know
I'm still playing a bit with mykid, but mostly I'm touching a
truly honest you know I'm stillplaying a bit with my kid, but
mostly I'm touching a bit.
You know pokemon go sometimesto still get a conversation with
my wife, you know, on gaming.
Speaker 1 (03:34):
That's good.
That's good.
Well, I'm, uh, really excitedfor this conversation and you
know I'm.
I was just thinking back as asI was making some notes for what
I thought we should talk about,and in 2007, I had the distinct
privilege and I really doconsider it to be a privilege to
(03:57):
be a part of a company, one ofthe early, early innovators of
what we call now OTT, and youknow, at the time it was
transactional VOD.
The company still exists, it'scalled Voodoo, and we had this
crazy idea to take theBlockbuster.
(04:17):
You know, those who have beenaround for a little while will
remember Blockbuster videostores in the US, other
countries, they had theequivalent and eventually, I
think, blockbuster did expandoutside the US.
But you'd go to the video store, you'd rent a disc, dvd and
(04:38):
then eventually Blu-ray, and youwould drive home so excited for
the family to join around theTV and watch it.
And I can remember how shockingit was to have built this
amazing experience where everytitle was in stock.
And those of us who rememberthe video store remember that
(05:00):
that was part of the challengeOn new release day, you had to
rush down to the store to be thefirst in line so you could even
get the movie because they onlyhad so many copies.
And then, of course, you had toworry about did I return it?
Did I return it by the deadlineor do I have to pay for a
second day?
There was a lot about theexperience that actually wasn't
(05:23):
so great, and yet we wereshocked.
Shocked, how many people saidwhy would I want to stream over
the internet like the like?
Dvd is great.
This is amazing.
Look at the quality.
No one's going to want toreplace the dvd.
Well, you know, 15 years later,obviously that sounds absolutely
(05:45):
crazy, as now the entire worldis streaming and we can't even
imagine a world without it.
But as I was thinking aboutcloud gaming, it feels like
maybe we're a little bit furtherthan we were in 2007, but
they're still not everybody'sconvinced, and I'm even
(06:07):
surprised that major publishersthat I'm coming across, and it's
not a foregone conclusion thatthe console is going to be
replaced with streaming.
And so let's start there.
Olivier, I have to imagine thata lot of what you're spending
(06:27):
time doing, aside from buildingthe technology, is making the
case for why, you know, internetdelivery of a game experience
is going to be better and isultimately better, you know,
than something that's installedon a PC, you know, downloaded,
or a console.
(06:48):
So what insights do you have toshare about where we are, you
know, in this transition fromconsoles and disks to streaming
for games?
Speaker 2 (07:01):
Yeah, and Mark, I think the analogy with, you know
, with the blockbusters, I think, is very relevant and I feel
that, first, in terms of marketmaturity for the end user, we
are probably at that point wherepeople would question why
should I do that?
I can download a game, whyshould I actually stream it?
Why do we need somethingdifferent?
(07:22):
Right, and when I createdJackknife, actually, a person
that I highly respect told mewell, people will not use it
because they can't download it.
Now, if you look at where we areright now, with people now
consuming all the media, likeaudio and video and your music,
and in a streaming manner, right, it seemed that definitely
(07:46):
having those people accessinggames the same way seems to be
actually the right idea or theright next step.
Right, and I do think thatthere is a bit more of maturity
of people actually willing toaccess games this way.
Now there has been probably aninflection point in terms of
technology maturity.
(08:07):
I think the technology meaningbasically the hardware you can
have on the cloud, the bandwidthyou have available on your home
as a kind of device you have torun it, and so on is good
enough to provide actually agreat experience, and I do think
(08:28):
that we are at the time herewhere we are passing this
inflection point that probablyyears ago it was not sufficient
and we have seen a lot ofcompanies trying to do this but
actually failing, and failingreally badly, but actually
learning a lot from thesefailures.
So I think we are at a veryexciting time now where we have
(08:48):
this maturity in terms oftechnology.
We have the maturity of the enduser, because they are used to
consume this kind of media withaudio video, e-books and so on,
so probably they are craving toget access to game and more and
more people are gaming.
And we have also the maturityof the content owner and the
publisher.
So I think we are at a very,very good time in the market.
Speaker 1 (09:11):
Well, I definitely agree that.
You know we are much furtheradvanced than we were, I think,
of some of the things that wehad to do.
Voodoo in 2007 actuallyrequired a, an appliance, a
device with a hard drive in itthat we could download the the
(09:34):
first, you know, 30 seconds,maybe a minute, of every single
library, every single title inthe library, and it at that time
the library was not as big asyou know what the libraries are
today, but just becausestreaming bandwidth was, you
know, was 768 kilobits, you know, maybe 1.5 megabits was like
(09:58):
really fast.
If you were really lucky, youhad five megabits.
My how we've grown.
So you know it's definitelywe're in a better position
Before we get into thetechnology, because that's where
we're going to spend the bulkof our time today.
But something that I think alsoyou're in a really good position
(10:19):
to address is the cost side.
Is is the cost side.
So certainly we're at a placetoday with the cloud, that you
can deliver anything reallyanywhere via the cloud.
So the notion that you can docloud gaming, ie it's possible
(10:39):
to deliver an ultra low latency,very high quality experience
from the cloud.
Deliver an ultra low latency,very high quality experience
from the cloud.
I don't think anybodyconceivably would say you know,
oh, I don't believe that that'snot possible.
But there is a real issue ofthe cost, and so why don't you
address where we're at in termsof just delivery cost, and I'm
(11:02):
speaking of, like OpEx.
You know where are we at.
I mean, is this possible butnot affordable?
Or is this possible andaffordable even for someone who
might not be able to chargetheir consumer a whole lot of
money?
You know, not all markets arethe US or Western Europe, or you
(11:25):
know some of these regionswhere consumers are willing to
pay, you know, $10, $15, $20 amonth.
Speaker 2 (11:34):
That's really the key issue, mark, because, as you
mentioned, I think we passed thetechnology inflection point
where actually the servicebecomes to be feasible.
Okay, technically feasible.
The experience is good.
We think it's good enough forthe mass market.
I am sure that some people willbe unhappy with it.
You know, really core gamers.
(11:54):
They say, well.
You know probably the samepeople that you know when the
DVD came.
They say, well, I still want tolisten to my vinyl on my rose
boot.
You know platine, because thisis what I'm want to listen to,
my vinyl on my rosebud.
You know, because this is whatI'm using to listen to my music.
And you will not beat thatquality with digital sound,
right, yeah, but for the massmarket, I think we got to the
point where the feasibility ishere, of course, we need good
(12:17):
bandwidth, stable, you know,very low jitter, so the
variation of the latency.
But we are here right Now.
The issue is indeed on the uniteconomics and how much it costs
to actually stream and delivergames in an efficient manner so
that it is affordable, basically, for the mass market.
(12:38):
And one thing here is I thinkthe game is not done.
Okay, there is some challenges,as you know.
The cost of streaming dependson the number of hours per month
.
Let's say that you stream.
We think that we got at least amaturity where it's becoming
available, so that you get to aprice point, which is what
(13:01):
people expect, which is, youknow, between $5 to $15,
depending on the harpy of thecountries.
So we think this is realistic,but of course it depends on the
intensity of the player, howmuch they play, and if you want
somehow to really sustain and tohave great economics, there is
still some improvement to bedone and I would say we have a
(13:25):
baseline architecture thatallows the service to be
profitable, to make it reallywork, really scale.
There is still some margin ofimprovement and we have ways
actually to improve this uniqueeconomics.
Speaker 1 (13:37):
So you're saying right now that to the end user
which means that the actual costto deliver the service has to
be, you know, less but to theend user, about $5 a month to
$15 a month is a target that ispossible to reach, right.
Okay, so $5 a month, you know,even in more emerging markets,
(14:04):
where maybe subscription pricescannot be what they are, say in
the US, feels like that's doable.
So that's actually good to hear.
Tell us what is the technical.
Let's talk now about what thetechnical infrastructure looks
like and what it takes todeliver.
(14:26):
How have you built your system?
And then we will get to thebroader architecture of
BlackNuts and what exactlyyou're offering.
But let's start with what isyour system built on?
What does it look like?
What are you deploying?
Is this a cloud service, youknow?
(14:47):
Is it run all on-prem?
Speaker 2 (14:49):
So, basically, the architecture of cloud gaming is
somehow simple.
Right, you take games, you putthem on a server in the cloud
and you are going basically tovirtualize it and stream it in
the form of a video stream or insome other format, you know, so
that you don't have to downloadthe game on the client side and
you can play it as you areplaying a video stream.
(15:09):
When you interact with a game,you send a comment back to the
server and then you interactwith the game.
This way, Of course, bandwidthneed to be sufficient, let's say
, six megabit per second.
Latency need to be good.
Let's say less than 80milliseconds Of per second
latency need to be good.
Let's say, you know, less than80 milliseconds.
And, of course, you need to havethe right infrastructure on the
server that can run games.
(15:31):
You know games mean a mixtureof CPU, GPU, storage, okay, and
all this needs to work well.
We start deploying, you know,the service based on public
cloud, because this allows us totest the different metrics how
people were playing the service,how many hours, and this was
(15:53):
actually very fast to launch andto scale.
So this is what the publiccloud, the hyperscaler, the
WSJCP and so on provides.
That's great, but they arequite expensive, right as you
know.
So to optimize, the economics weactually built and invented in
BlackNet what we call the hybridcloud for cloud gaming, which
(16:15):
is a combination of both thepublic cloud and private cloud.
So we have to install our ownservers, you know, based on GPUs
, cpus and so on, eitherdirectly in Black Nut or with
some partners like Radiant Arc,so that we can improve the
overall performances and theunit economics of the system.
(16:35):
That, I think, allowed us tobuild a profitable service.
I think, if you just matchbasically the public cloud,
currently I think this is superhard to get something which is
viable, but with this kind ofhybrid cloud I think it's
actually very doable.
Speaker 1 (16:52):
And these are standard x86, commercial
off-the-shelf Intel AMD machines.
I mean there's nothing specialrequired, or have you gone to a
purpose-built design?
Speaker 2 (17:08):
No, the current design is basically definitely
specific for the private cloud,but it's based on standard x86.
And for GPU we use AMD orNVIDIA.
We have a mixture of differentproviders, but basically this is
, I would say, reasonablystandard architecture with a mix
(17:30):
of CPU, gpu and storage.
Speaker 1 (17:33):
The cloud gaming use case is a primary one and that's
obviously why we got introducedand you are using NetEnt, which
we will get to.
But kind of the key measure froma technology perspective and it
maps directly back to cost fora cloud gaming installation is
(17:54):
the number of concurrentsessions per server.
Obviously, it just stands toreason that the more concurrent
sessions or players that you canget on a server, well, it's
going to be less expensive tooperate and to run, so that's
not too difficult to understand.
One of the things that's reallyinteresting is and I'd like for
(18:16):
you to talk about thisarchitecture where you have the
GPU rendering the game butyou're actually not doing the
video encoding on the GPU.
So what does that look like?
And also talk to us about.
You know the evolution, becausethat's not where you started
(18:38):
and most cloud gaming platformstoday are attempting to keep
everything on the GPU, which hassome advantages, but it has
some very distinct disadvantagesand trade-offs, which means
that your cost per stream likelycannot meet that economic bar
(19:03):
where you can really affordablydeliver to a wider number of
players.
Ie, you can't drive your costsdown, so you have to charge more
and there's people who will say, well, that's too expensive,
but talk to us about thisarchitecture.
So that's correct, mark.
Speaker 2 (19:21):
I think the ultimate measure is the cost per CCU
right, the cost per current userthat you can get on a specific
bill of material right.
If you have a CPU plus GPUarchitecture, the game is going
to actually slice the GPU indifferent pieces in the more
(19:42):
dynamic manner and in the moreappropriate manner so that you
can run different games and asmuch game as possible.
So, typically, if you get onthe standard GPU, you can run
probably a big game like a largegame, and you can cut the GPU
in four pieces.
If you run a medium game like alarge game and you can cut the
GPU in four pieces If you run amedium game, you can run it
(20:05):
maybe in six or eight pieces andif you run a smaller game, then
maybe you can get to I don'tknow 20 pieces.
Right, there is some limits onhow much you can slice a GPU for
the GPU to be still efficientand likely.
You know, for example, theNVIDIA license allows you to
slice one GPU in 24 pieces, butthat's it, right, and so there
(20:29):
is some limits in thisarchitecture because it all
relies on the GPU.
We are indeed investigatingdifferent architecture where,
indeed, we are using a VPU right, like NetEat, is providing a
video processor that willsomehow offload the GPU of the
task of, you know, encoding andstreaming the video so that we
(20:51):
can augment the density.
Okay, and we see it, you know,has in terms of full
architecture as something whichwill be a bit more flexible
architecture as something whichwill be a bit more flexible, I
think, in terms of number of biggames, because they rely much
more on the GPU.
Probably you will not augmentthe densities that much, but we
(21:13):
think that, overall, probably wecan gain a factor of 10 on the
number of games that you can,overall, run on this kind of
architecture.
So, passing from a max of 20,24 games to a time 10, right,
running 200 games on anarchitecture of this kind, yeah,
that's really remarkable.
Speaker 1 (21:29):
And just in case somebody isn't doing the quick
math here, what you're saying isis that with this CPU plus GPU
plus VPU which the VPU is theASIC based video encoder all in
(21:50):
the same chassis, so the sameserver we're not talking about
different servers you can get upto 200 game players
simultaneously, so concurrentplayers, which just radically
changes the economics.
And in our experience workingwith publishers and working with
(22:11):
platforms cloud gamingplatforms nearly everybody has
said literally, without thatit's not even really economical
to build the platform, you know.
In other words, you end uphaving to charge your customers
so much and where the experienceis, it's not viable.
Speaker 2 (22:34):
So that's important and for certain category of
games, definitely you can reachyou know this level and for
certain category of games,definitely you can reach this
level.
So actually augmenting thedensity by a factor of 10 means
also, of course, diminishing thecost per CCU by a factor of 10.
So if you pay $1, currently youwill pay $0.10.
And that makes a wholedifference because, let's assume
(22:58):
, basic gamers will pay 10 hoursper month or 30 hours per month
.
If this is $1, this is $30,right, if this is $10, then you
go to $1 to $3, which I thinkmakes the math work.
Speaker 1 (23:10):
On the subscription, which is, between you know, 5 to
15 euros per month, One of thequestions that comes up and I
know you know we've had thisconversation with you is how is
this possible?
Because anybody who understandsbasic server architecture you
(23:30):
know.
Basically it's not difficult tothink well, wait a second,
isn't there a bottleneck insidethe machine?
And this must require a reallysuper hot, rotted machine, you
know.
So maybe the cost savings isoffset by super expensive
hardware, and I think it'simportant to note that the
(23:53):
reason why this is possible is,first of all, the VPU is built
on NVMe architecture, so it'susing the exact same storage
protocol as your hard drive, youknow, as the SSDs that are in
the machine.
And what we have done, whatNetEnt has done, is actually
(24:17):
created a peer-to-peer sharinginside the DMA.
So, basically, the GPU willoutput a frame, a rendered frame
, and it's transferred literallyinside memory so that then the
VPU can pick that up, encode itand there's effectively zero
(24:39):
latency, at least in terms ofthe latency is so low because
it's happening in the memorybuffer.
And so if anybody's listeningand raising an eyebrow,
wondering, well, wait a second,you know, surely there's a
bottleneck, and especially ifyou're talking 60 frames per
second which, by the way, ourbenchmarks are generally always
(25:01):
at 60 frames per second because,you know, unless it's real
casual games, you need thatframe rate to really deliver a
great experience, even aboveresolution, in some cases it's
better to get the frame rate upthan to increase the size of the
frame.
Absolutely, absolutely,absolutely.
(25:21):
Yeah, let me just pause here andsay that we would love to have
questions, and so feel free, onwhatever platform you're if
you're on YouTube or LinkedIn orwherever you're watching us
right now just type in and Iwill try and pick those up.
(25:42):
I have looks like we alreadyhave one, so you know why don't
we?
I think this is actually areally good one.
I'm gonna pick this up righthere, but feel free to enter
questions in the chat.
So, Olivier, the question is Ilive in a country where stable
(26:08):
internet is not always availableand, by the way, I would say
that you know this isn't only,you know, a country issue.
Internet varies, right, and theexpectation of users is more
and more that.
You know they don't think aboutthe fact that I'm in a car.
You know I happen to be in anarea where there's great
(26:29):
coverage, but you know, sevenmiles down the road that changes
right.
They want to keep playing andkeep enjoying this great
experience.
So the question is I live in acountry where stable internet is
not always available.
How will this affect the gamingexperience?
And you know, yeah, I mean,that's the question.
(26:52):
So, what's your experience andhow are you guys solving for
this?
Speaker 2 (26:56):
You see, in Netflix or Spotify, right, you can
actually buffer content so that,even if your bandwidth is, you
know, a bit clumsy, you canactually store that content in a
CDM and keep the experience,you know good enough, right, or
you can download the thatcontent in a CDN and keep the
experience good enough, or youcan download the video and make
it work.
So, definitely you have someway to solve that problem in, I
(27:19):
would say, cold media, mediathat you can encode in one way,
then stream later.
In games, this is completelydifferent.
Yeah, you can't do that Becausewe have to encode, stream,
deliver and then you take theiteration right away.
So if your bandwidth is notenough, right.
If the quality of the bandwidthis not enough, and not only in
(27:41):
terms of the size of thebandwidth but also in terms of
characteristics, the latency,how this latency is stable, and
so on, then the experience willbe great.
So what we've been doingactually with Ericsson is to use
5G networks and to definespecific characteristics of what
(28:01):
is a slice in the 5G network,so we can tune the 5G network to
make it fit for gaming and tooptimize, basically, the
delivery of gaming with 5G.
So we think that 5G is going toget much faster in those
regions where actually theinternet is not so great.
We've been deploying theBlacknet service in Thailand, in
(28:27):
Singapore, in Malaysia, now inthe Philippines and so on, and
this has allowed us to actuallyreach people in regions where
there is no cable or bandwidthwith fiber and this kind of
thing.
So, look, I'm not going tosolve the problem where
(28:47):
bandwidth is not available, butmaybe bandwidth will come faster
with 5G and that could be thesolution.
Speaker 1 (28:54):
Yeah, I want to make a comment there and thank you
for the answer we are seeing.
So it's very interesting andI'll use India as an example.
So for years in video streaming, the Indian market was used as
an example of where it was verydifficult to deliver high
(29:19):
quality and especially if youwanted to deliver, say, 720p and
1080p was almost assumed at acertain period of time like it's
not even possible because thenetwork capacity and the speeds
were just so low.
What has happened is and Indiais a great case study here, but
(29:43):
it's really almost all regionsof the world as these
infrastructures, these wirelessinfrastructures, have been
upgraded, they leapfroggedliterally from 3G or you know in
some cases even you know 2.5Gand you know before and just
went all the way to 5G.
(30:03):
And so in the last five yearsthere has been such a
fundamental shift in bandwidthavailability that in some cases,
some of these regions of theworld, not only is it definitely
no longer true that they'reslow, they're faster than some
(30:23):
of the more developed countries.
So I do want to make thatstatement there.
One question, olivier, can youtalk about?
Is this WebRTC?
What protocols you're using?
There's a lot of talk right nowabout QUIC and I think that
(30:45):
would be interesting for some ofthe listeners who might be
wondering even what protocolsyou're using.
Speaker 2 (30:51):
So we use standard codecs to start with the bottom
line.
We have not invented codecs.
We have been in thestandardization industry of
audio and video for quite someyears and I think you have great
experts here doing greattechnology.
And this technology is actuallyembedded into the chipset, into
the hardware, so actually youcan rely on hardware encoding
(31:15):
and decoding capabilities.
So we do think standard codecsis basically a must-have, right?
Of course, you need toconfigure them the right way,
because you have to codereal-time, okay, so you cannot
use particular techniques.
You know to wait for a coupleof frames to compute more, so
you have to optimize this, butbasically we use standard
(31:35):
collects.
Then, on the protocols, on topof this, we have actually a
large variety of protocol okay,we depend on the device on which
you are streaming.
So it goes from fullproprietary protocol that we
have invented and patented inBlackNut to standard WebRTC.
So, if you look at devices likeSamsung and LG, which are
(32:00):
basically the top manufacturers,I think the service has been
launched on LG we are going toannounce, I think, our launch
with Samsung in a very shorttime and these devices support
WebRTC, and that's basically theonly way to implement and to
support the cloud gamingsolution efficiently.
(32:22):
So short answer we use a widerange of protocols, always the
one that is the most appropriateand provides the best
experience to the end-user.
We are using that, of coursenew protocol, new standards,
experimenting this but I wouldsay for the main streamline new
solution, we use our ownsolution, plus WebRTC, when this
(32:45):
is the only standard that therethe end-to-end latency targets.
Speaker 1 (32:50):
I think previously you made the comment about you
know like 80 milliseconds, butgive us some guidelines.
What is obviously the answer isas low as possible, but what's
the upper limit where the gameexperience just falls apart?
You know, it's just notplayable.
Speaker 2 (33:09):
You know that the limit for conversional video is
about 150 milliseconds.
For playing games this is muchlower, probably half of it.
So I think you know you can geta reasonably good experience at
80 milliseconds, or actuallymost of the games.
That does not require you knowthis kind of fast reaction.
But then if you want to go toFPS or this kind of thing that
(33:32):
really need to nearly bereactive at the frame accuracy,
which is very, of course,difficult in cloud gaming, you
need to go down to the 30milliseconds and lower.
And that I think is onlyfeasible if you have the network
that allows for it, becauseit's not only about the encoding
part, the server side, and theclient side, it's also on where
(33:57):
the packets are going throughthe networks, because you can
have the most efficient systemsin terms of encoding latency and
decoding latency.
But if you pocket, instead ofgoing directly from the server
to the end user, go here andthere and transit in many places
, your experience will be crappy.
And you know, Mark, this isactually a real issue because we
(34:18):
, for example, had a greatdemonstration with Ericsson in
Madrid, you know, in Barcelona,right for the Mobile World
Congress, and we had servers inMadrid for the Mobile World
Congress, and we had servers inMadrid.
But when we first made thefirst test, we discovered that
the packet was going from Madridto Paris and back to Barcelona,
right, okay.
So we needed a bit ofintelligence and technology to
(34:41):
make this connection asefficient as possible.
Speaker 1 (34:44):
Tell us about Black Nut.
What exactly do you guysdeliver?
Speaker 2 (34:50):
We provide basically a cloud gaming service which is
let's categorize it as a game asa service.
This means that for thesubscription fee per month, you
get access to the real stuff.
You get access to 700 games.
We are adding 10 to 15 newgames per month, which is, I
think, the fastest pace in termsof increasing games on the
(35:12):
market, and we provide thisexperience on all single devices
that can actually receive avideo.
So that's what we do and wedistribute this service either
B2C, so direct to the consumer.
So if you go on your BlackNetwebpage, you know you can
subscribe, you can access to thegames.
But we also distributed throughcarriers, so telecommunication
(35:36):
carriers, operators all over theworld.
We currently have about 20signed agreements with the
carriers live actually more than40 signed and we are signing
and delivering one to two newcarriers per month, you know.
So that basically is 40 signand we are signing and
delivering one to two newcarriers per month.
So that's basically the pacewhere we are in BACnet and the
(35:56):
choice to use carriers here isfor the reason I explained to
you, that it's good to havegreat service.
Speaker 1 (36:01):
Optimization of the network right.
Speaker 2 (36:05):
You need to know where the packets are going.
You need to make sure thatthere is some form of CDN for
cloud gaming that is in placehere that makes the experience
optimal.
Speaker 1 (36:14):
Yeah, it completely makes sense to me, especially
because you mentioned the 5Goptimization.
And obviously, carriers.
They've been investing now foryears in building out their 5G
networks but they're alwayslooking for reasons to drive
more value and to really extractthe full potential off the 5G
(36:36):
or out of the 5G investment.
So, yeah, it really makes sense.
Speaker 2 (36:41):
That's the kind of thing we are doing as well with,
you know, our partner,radianark, and we are putting
the server at the edge of thenetwork.
So inside you know thecarrier's infrastructure, so
that you know the latency isreally super optimized.
So that's one thing that is keyfor the service.
Speaker 1 (36:57):
What is the architecture of that edge server
?
What's in it?
What CPU, GPU, VPU, like?
Describe that.
Speaker 2 (37:07):
We started with a standard architecture, you know,
with CPU and GPU, and now, youknow, with the current you know
VPU architecture, we are puttingactually a whole server
consisting in AMD GPU, you knowNetIn GPU, and basically we
build the whole package so thatwe put this in the
(37:27):
infrastructure of the carrierand we can deploy the BlackBerry
Cloud Gaming on top of it.
Speaker 1 (37:33):
And are you delivering to only a handful of
fixed resolutions, or like if Iwas on a TV, for example, do I
get 4K, or do you limit to 1080p, or how do you handle that?
Again, great question.
Speaker 2 (37:52):
We actually can handle multiple resolution,
right.
I think we can stream from 720pup to 4k.
Uh, the technology basicallyhas no limits for it, right?
Um, and streaming you know 4kor even you know 8K is a problem
that has somehow been solvedalready.
From a technical matter.
The question is again the costand the experience.
(38:16):
Streaming 4K on a mobile devicedoes not really make sense.
I think the screen is a bitmore, so you can screen a
smaller resolution and that'ssufficient.
On a TV, likely you need tohave a bigger resolution, even
if actually there is a greatupscale available on most of the
(38:38):
TV sets.
We stream 720p on Samsungdevices and that's super great,
right, but of course, scanningup to 1080p will provide a much
better experience.
So, on TVs and for the gamethat require it, I think we are
indeed streaming the serviceabout 1080p for the game that
(39:00):
requires it.
Speaker 1 (39:01):
Do you also find that frame rate is almost more
important than resolution Forcertain?
Speaker 2 (39:08):
games, absolutely, but you know again, it is game
dependent.
It's game.
Yeah, if you are on FPS, youprobably.
If you have the choice, okay,and you cannot stream 1080p, you
would probably stream 720p at60 FPS rather than 1080p at 30
FPS, right?
Yes, yes, if you have to makesome trade-off, but if you have
(39:30):
different games where thetextures, the resolution is more
important, then maybe you willactually select more 1080p and
30fps resolution and thetechnology we build is actually
fully adaptable.
Ultimately, you should notforget that there is a network
in between and even if,technically, you can stream 4K
(39:51):
or 8K, the networks may notsustain it and then actually you
will have less good experiencestreaming 4K than actually a
1080p 60fps resolution.
Speaker 1 (40:04):
Okay, I see, a question just came in
(40:28):
no-transcript.
You actually talked about howmany publishers you have.
You did talk about every monthyou're onboarding, you know, I
think, 10 or 12 new games, butyeah, so are there geographical
restrictions?
How can someone access this?
Speaker 2 (40:48):
great, let's start's start with content.
Okay, indeed, we have, you know, more than several games right
now, 10 to 15 new games permonth, and we actually try not
to have geographical limitationon the content.
Okay, so it's been.
The content we have on thecatalog is, from a licensing
point of view, availableworldwide.
(41:10):
That's basically what we do.
We do have exceptions, as usual, but basically a large part of
the catalog is availableworldwide.
Now, deploying this catalog ondifferent region, we are
available in more than 45countries.
We definitely need to haveservers that are close enough to
(41:32):
the end user so that thestreaming experience is good
enough, and we think that aradius of between 750 to 1,500
kilometers probably is themaximum kilometers, you know,
probably the maximum.
So I think we will actually putsome point of presence, you
(41:57):
know, in those geographicalareas so that, basically, the
latency, limited by the speed oflight right, does not, you know
, harm the service.
So, of course, you know, if youlook at it, we have europe very
much, you know, covered.
We have us and canada very muchcovered.
We have a large portion ofsoutheast asia, korean and japan
(42:18):
very much covered.
We are now expanding in latinamerica, which is a bit harder.
We are strong presence now aswell in in the middle with
partners like STC in the regionand of course we have some zones
that are less covered.
Africa is not well covered atall.
(42:39):
South Africa is, but basicallythe rest of Africa is a bit
harder to reach.
Speaker 1 (42:45):
By the way, what is the website?
Why don't you give out the?
Speaker 2 (42:50):
URL there wwwblacknutcom.
Speaker 1 (42:54):
Blacknutcom, all right.
Speaker 2 (42:57):
I think, try the service.
We'd be very happy to supportand provide feedback.
Speaker 1 (43:05):
It's super exciting and you know, as I, as I said in
the in the beginning, for mepersonally, having been, you
know, really, in the very earlystages of of the transition from
physical entertainment deliveryI'm talking about movies, you
know, specifically like DVDs tostreaming.
(43:26):
I'm just super excited to alsonow, 15 years later, be there
with games.
And there's a lot of work to bedone and, as you pointed out,
you know, the experience isabsolutely not exactly mapped.
We can't replace, you know,throw out the console yet.
(43:47):
You know, throw out the consoleyet, uh, and but the, the
opportunity to bring, you know,really, the gaming experience to
, um, to a much wider audience,is really enabled, you know,
with streaming.
So, um, by by the way, uh.
So I think there's a follow-onquestion here Do you have
(44:08):
infrastructure in South Africayou mentioned?
Africa is not covered as well.
Speaker 2 (44:13):
Yes, we do have the capacity to deploy the service
in South Africa.
Speaker 1 (44:18):
To deploy in South Africa.
Okay, great, great.
Well, we're right up againsttime and thank you for everyone
who joined us.
Live Really appreciate it.
And thank you, Olivier, it'samazing what you built and we're
super excited to be workingwith Black Nut.
Thank you everyone.
Speaker 2 (44:36):
Thanks, Mark.
This episode of Voices of Videois brought to you by NetInt
Technologies.
If you are looking for cuttingedge video encoding solutions,
check out NetInt's products atnetintcom.
Voices of Video.
So we are at the top of thehour and looks like we should
get started.
Olivier, are you ready to talkabout cloud gaming?
(00:29):
Absolutely ready, excellent,excellent.
Well, welcome to those who arejoining us live.
This is the May edition ofVoices of Video, and, if you
haven't joined us before, voicesof Video and if you haven't
joined us before, voices ofVideo is a conversation, or,
(00:50):
some might say, a real dialogue,not a podcast, I guess, a video
cast.
We go live on LinkedIn and alsoa lot of other platforms, and
we are talking each month withinnovators in the video space,
and so this month, I am superexcited to have Olivier Avaro,
(01:12):
who is the CEO of a companycalled Black Nut, and we are
talking about cloud gaming.
I will let Olivier tell us allabout what his company does, but
welcome to Voices of Video,olivier.
Speaker 2 (01:28):
Thanks a lot, mark, for the nice introduction.
So my name is Olivier Varro.
I'm the CEO of BlackNerd, which, in short, is doing two games,
what Spotify did for music, sowe are distributing games from
the cloud large, large catalogof games more than 700 games so
(01:50):
far and this, you know, for asimple subscription fee, right?
Uh, I was long time a gamer.
I enjoyed it a lot, you know,when I was a teenager.
I enjoyed a lot with friends,with my family, later with my
kids, and I started Black Nut in2016 with the big ambition to
actually bring this joy ofgaming, this good emotion, all
(02:12):
the positive value of, you know,playing together, to the mass
market.
We deployed the tech for aboutthree years, I think cloud
gaming is, you know, doesrequire a bit of technology to
work efficiently.
Then we started deploying itall over the world, and this is
where we are today.
Speaker 1 (02:30):
I love it.
So I have to ask the question.
You know, sometimes when we'rebuilding advanced technologies
we get so into the technology wedon't get to do the thing that
we originally set out to do,like play games.
So are you still a gamer?
Well, I'm you set aside timeeach day to play or I set aside
(02:52):
each time to play a little bit.
Speaker 2 (02:53):
That's true.
And I have to say that, youknow I was the first game I
played was, you know, on theCommodore 64 machine.
It was then Bulldash, right, sothe older of the audience will
know about it.
Now I'm still.
I've been playing with my kid,you know, of course, on the Wii.
You know all the Nintendo gamesand Mario and Super Mario Kart
(03:18):
and Super Mario Galaxy, right,and to be truly honest, you know
I'm still playing a bit with mykid, but mostly I'm touching a
truly honest you know I'm stillplaying a bit with my kid, but
mostly I'm touching a bit.
You know pokemon go sometimesto still get a conversation with
my wife, you know, on gaming.
Speaker 1 (03:34):
That's good.
That's good.
Well, I'm, uh, really excitedfor this conversation and you
know I'm.
I was just thinking back as asI was making some notes for what
I thought we should talk about,and in 2007, I had the distinct
privilege and I really doconsider it to be a privilege to
(03:57):
be a part of a company, one ofthe early, early innovators of
what we call now OTT, and youknow, at the time it was
transactional VOD.
The company still exists, it'scalled Voodoo, and we had this
crazy idea to take theBlockbuster.
(04:17):
You know, those who have beenaround for a little while will
remember Blockbuster videostores in the US, other
countries, they had theequivalent and eventually, I
think, blockbuster did expandoutside the US.
But you'd go to the video store, you'd rent a disc, dvd and
(04:38):
then eventually Blu-ray, and youwould drive home so excited for
the family to join around theTV and watch it.
And I can remember how shockingit was to have built this
amazing experience where everytitle was in stock.
And those of us who rememberthe video store remember that
(05:00):
that was part of the challengeOn new release day, you had to
rush down to the store to be thefirst in line so you could even
get the movie because they onlyhad so many copies.
And then, of course, you had toworry about did I return it?
Did I return it by the deadlineor do I have to pay for a
second day?
There was a lot about theexperience that actually wasn't
(05:23):
so great, and yet we wereshocked.
Shocked, how many people saidwhy would I want to stream over
the internet like the like?
Dvd is great.
This is amazing.
Look at the quality.
No one's going to want toreplace the dvd.
Well, you know, 15 years later,obviously that sounds absolutely
(05:45):
crazy, as now the entire worldis streaming and we can't even
imagine a world without it.
But as I was thinking aboutcloud gaming, it feels like
maybe we're a little bit furtherthan we were in 2007, but
they're still not everybody'sconvinced, and I'm even
(06:07):
surprised that major publishersthat I'm coming across, and it's
not a foregone conclusion thatthe console is going to be
replaced with streaming.
And so let's start there.
Olivier, I have to imagine thata lot of what you're spending
(06:27):
time doing, aside from buildingthe technology, is making the
case for why, you know, internetdelivery of a game experience
is going to be better and isultimately better, you know,
than something that's installedon a PC, you know, downloaded,
or a console.
(06:48):
So what insights do you have toshare about where we are, you
know, in this transition fromconsoles and disks to streaming
for games?
Speaker 2 (07:01):
Yeah, and Mark, I think the analogy with, you know
, with the blockbusters, I think, is very relevant and I feel
that, first, in terms of marketmaturity for the end user, we
are probably at that point wherepeople would question why
should I do that?
I can download a game, whyshould I actually stream it?
Why do we need somethingdifferent?
(07:22):
Right, and when I createdJackknife, actually, a person
that I highly respect told mewell, people will not use it
because they can't download it.
Now, if you look at where we areright now, with people now
consuming all the media, likeaudio and video and your music,
and in a streaming manner, right, it seemed that definitely
(07:46):
having those people accessinggames the same way seems to be
actually the right idea or theright next step.
Right, and I do think thatthere is a bit more of maturity
of people actually willing toaccess games this way.
Now there has been probably aninflection point in terms of
technology maturity.
(08:07):
I think the technology meaningbasically the hardware you can
have on the cloud, the bandwidthyou have available on your home
as a kind of device you have torun it, and so on is good
enough to provide actually agreat experience, and I do think
(08:28):
that we are at the time herewhere we are passing this
inflection point that probablyyears ago it was not sufficient
and we have seen a lot ofcompanies trying to do this but
actually failing, and failingreally badly, but actually
learning a lot from thesefailures.
So I think we are at a veryexciting time now where we have
(08:48):
this maturity in terms oftechnology.
We have the maturity of the enduser, because they are used to
consume this kind of media withaudio video, e-books and so on,
so probably they are craving toget access to game and more and
more people are gaming.
And we have also the maturityof the content owner and the
publisher.
So I think we are at a very,very good time in the market.
Speaker 1 (09:11):
Well, I definitely agree that.
You know we are much furtheradvanced than we were, I think,
of some of the things that wehad to do.
Voodoo in 2007 actuallyrequired a, an appliance, a
device with a hard drive in itthat we could download the the
(09:34):
first, you know, 30 seconds,maybe a minute, of every single
library, every single title inthe library, and it at that time
the library was not as big asyou know what the libraries are
today, but just becausestreaming bandwidth was, you
know, was 768 kilobits, you know, maybe 1.5 megabits was like
(09:58):
really fast.
If you were really lucky, youhad five megabits.
My how we've grown.
So you know it's definitelywe're in a better position
Before we get into thetechnology, because that's where
we're going to spend the bulkof our time today.
But something that I think alsoyou're in a really good position
(10:19):
to address is the cost side.
Is is the cost side.
So certainly we're at a placetoday with the cloud, that you
can deliver anything reallyanywhere via the cloud.
So the notion that you can docloud gaming, ie it's possible
(10:39):
to deliver an ultra low latency,very high quality experience
from the cloud.
Deliver an ultra low latency,very high quality experience
from the cloud.
I don't think anybodyconceivably would say you know,
oh, I don't believe that that'snot possible.
But there is a real issue ofthe cost, and so why don't you
address where we're at in termsof just delivery cost, and I'm
(11:02):
speaking of, like OpEx.
You know where are we at.
I mean, is this possible butnot affordable?
Or is this possible andaffordable even for someone who
might not be able to chargetheir consumer a whole lot of
money?
You know, not all markets arethe US or Western Europe, or you
(11:25):
know some of these regionswhere consumers are willing to
pay, you know, $10, $15, $20 amonth.
Speaker 2 (11:34):
That's really the key issue, mark, because, as you
mentioned, I think we passed thetechnology inflection point
where actually the servicebecomes to be feasible.
Okay, technically feasible.
The experience is good.
We think it's good enough forthe mass market.
I am sure that some people willbe unhappy with it.
You know, really core gamers.
(11:54):
They say, well.
You know probably the samepeople that you know when the
DVD came.
They say, well, I still want tolisten to my vinyl on my rose
boot.
You know platine, because thisis what I'm want to listen to,
my vinyl on my rosebud.
You know, because this is whatI'm using to listen to my music.
And you will not beat thatquality with digital sound,
right, yeah, but for the massmarket, I think we got to the
point where the feasibility ishere, of course, we need good
(12:17):
bandwidth, stable, you know,very low jitter, so the
variation of the latency.
But we are here right Now.
The issue is indeed on the uniteconomics and how much it costs
to actually stream and delivergames in an efficient manner so
that it is affordable, basically, for the mass market.
(12:38):
And one thing here is I thinkthe game is not done.
Okay, there is some challenges,as you know.
The cost of streaming dependson the number of hours per month
.
Let's say that you stream.
We think that we got at least amaturity where it's becoming
available, so that you get to aprice point, which is what
(13:01):
people expect, which is, youknow, between $5 to $15,
depending on the harpy of thecountries.
So we think this is realistic,but of course it depends on the
intensity of the player, howmuch they play, and if you want
somehow to really sustain and tohave great economics, there is
still some improvement to bedone and I would say we have a
(13:25):
baseline architecture thatallows the service to be
profitable, to make it reallywork, really scale.
There is still some margin ofimprovement and we have ways
actually to improve this uniqueeconomics.
Speaker 1 (13:37):
So you're saying right now that to the end user
which means that the actual costto deliver the service has to
be, you know, less but to theend user, about $5 a month to
$15 a month is a target that ispossible to reach, right.
Okay, so $5 a month, you know,even in more emerging markets,
(14:04):
where maybe subscription pricescannot be what they are, say in
the US, feels like that's doable.
So that's actually good to hear.
Tell us what is the technical.
Let's talk now about what thetechnical infrastructure looks
like and what it takes todeliver.
(14:26):
How have you built your system?
And then we will get to thebroader architecture of
BlackNuts and what exactlyyou're offering.
But let's start with what isyour system built on?
What does it look like?
What are you deploying?
Is this a cloud service, youknow?
(14:47):
Is it run all on-prem?
Speaker 2 (14:49):
So, basically, the architecture of cloud gaming is
somehow simple.
Right, you take games, you putthem on a server in the cloud
and you are going basically tovirtualize it and stream it in
the form of a video stream or insome other format, you know, so
that you don't have to downloadthe game on the client side and
you can play it as you areplaying a video stream.
(15:09):
When you interact with a game,you send a comment back to the
server and then you interactwith the game.
This way, Of course, bandwidthneed to be sufficient, let's say
, six megabit per second.
Latency need to be good.
Let's say less than 80milliseconds Of per second
latency need to be good.
Let's say, you know, less than80 milliseconds.
And, of course, you need to havethe right infrastructure on the
server that can run games.
(15:31):
You know games mean a mixtureof CPU, GPU, storage, okay, and
all this needs to work well.
We start deploying, you know,the service based on public
cloud, because this allows us totest the different metrics how
people were playing the service,how many hours, and this was
(15:53):
actually very fast to launch andto scale.
So this is what the publiccloud, the hyperscaler, the
WSJCP and so on provides.
That's great, but they arequite expensive, right as you
know.
So to optimize, the economics weactually built and invented in
BlackNet what we call the hybridcloud for cloud gaming, which
(16:15):
is a combination of both thepublic cloud and private cloud.
So we have to install our ownservers, you know, based on GPUs
, cpus and so on, eitherdirectly in Black Nut or with
some partners like Radiant Arc,so that we can improve the
overall performances and theunit economics of the system.
(16:35):
That, I think, allowed us tobuild a profitable service.
I think, if you just matchbasically the public cloud,
currently I think this is superhard to get something which is
viable, but with this kind ofhybrid cloud I think it's
actually very doable.
Speaker 1 (16:52):
And these are standard x86, commercial
off-the-shelf Intel AMD machines.
I mean there's nothing specialrequired, or have you gone to a
purpose-built design?
Speaker 2 (17:08):
No, the current design is basically definitely
specific for the private cloud,but it's based on standard x86.
And for GPU we use AMD orNVIDIA.
We have a mixture of differentproviders, but basically this is
, I would say, reasonablystandard architecture with a mix
(17:30):
of CPU, gpu and storage.
Speaker 1 (17:33):
The cloud gaming use case is a primary one and that's
obviously why we got introducedand you are using NetEnt, which
we will get to.
But kind of the key measure froma technology perspective and it
maps directly back to cost fora cloud gaming installation is
(17:54):
the number of concurrentsessions per server.
Obviously, it just stands toreason that the more concurrent
sessions or players that you canget on a server, well, it's
going to be less expensive tooperate and to run, so that's
not too difficult to understand.
One of the things that's reallyinteresting is and I'd like for
(18:16):
you to talk about thisarchitecture where you have the
GPU rendering the game butyou're actually not doing the
video encoding on the GPU.
So what does that look like?
And also talk to us about.
You know the evolution, becausethat's not where you started
(18:38):
and most cloud gaming platformstoday are attempting to keep
everything on the GPU, which hassome advantages, but it has
some very distinct disadvantagesand trade-offs, which means
that your cost per stream likelycannot meet that economic bar
(19:03):
where you can really affordablydeliver to a wider number of
players.
Ie, you can't drive your costsdown, so you have to charge more
and there's people who will say, well, that's too expensive,
but talk to us about thisarchitecture.
So that's correct, mark.
Speaker 2 (19:21):
I think the ultimate measure is the cost per CCU
right, the cost per current userthat you can get on a specific
bill of material right.
If you have a CPU plus GPUarchitecture, the game is going
to actually slice the GPU indifferent pieces in the more
(19:42):
dynamic manner and in the moreappropriate manner so that you
can run different games and asmuch game as possible.
So, typically, if you get onthe standard GPU, you can run
probably a big game like a largegame, and you can cut the GPU
in four pieces.
If you run a medium game like alarge game and you can cut the
GPU in four pieces If you run amedium game, you can run it
(20:05):
maybe in six or eight pieces andif you run a smaller game, then
maybe you can get to I don'tknow 20 pieces.
Right, there is some limits onhow much you can slice a GPU for
the GPU to be still efficientand likely.
You know, for example, theNVIDIA license allows you to
slice one GPU in 24 pieces, butthat's it, right, and so there
(20:29):
is some limits in thisarchitecture because it all
relies on the GPU.
We are indeed investigatingdifferent architecture where,
indeed, we are using a VPU right, like NetEat, is providing a
video processor that willsomehow offload the GPU of the
task of, you know, encoding andstreaming the video so that we
(20:51):
can augment the density.
Okay, and we see it, you know,has in terms of full
architecture as something whichwill be a bit more flexible
architecture as something whichwill be a bit more flexible, I
think, in terms of number of biggames, because they rely much
more on the GPU.
Probably you will not augmentthe densities that much, but we
(21:13):
think that, overall, probably wecan gain a factor of 10 on the
number of games that you can,overall, run on this kind of
architecture.
So, passing from a max of 20,24 games to a time 10, right,
running 200 games on anarchitecture of this kind, yeah,
that's really remarkable.
Speaker 1 (21:29):
And just in case somebody isn't doing the quick
math here, what you're saying isis that with this CPU plus GPU
plus VPU which the VPU is theASIC based video encoder all in
(21:50):
the same chassis, so the sameserver we're not talking about
different servers you can get upto 200 game players
simultaneously, so concurrentplayers, which just radically
changes the economics.
And in our experience workingwith publishers and working with
(22:11):
platforms cloud gamingplatforms nearly everybody has
said literally, without thatit's not even really economical
to build the platform, you know.
In other words, you end uphaving to charge your customers
so much and where the experienceis, it's not viable.
Speaker 2 (22:34):
So that's important and for certain category of
games, definitely you can reachyou know this level and for
certain category of games,definitely you can reach this
level.
So actually augmenting thedensity by a factor of 10 means
also, of course, diminishing thecost per CCU by a factor of 10.
So if you pay $1, currently youwill pay $0.10.
And that makes a wholedifference because, let's assume
(22:58):
, basic gamers will pay 10 hoursper month or 30 hours per month
.
If this is $1, this is $30,right, if this is $10, then you
go to $1 to $3, which I thinkmakes the math work.
Speaker 1 (23:10):
On the subscription, which is, between you know, 5 to
15 euros per month, One of thequestions that comes up and I
know you know we've had thisconversation with you is how is
this possible?
Because anybody who understandsbasic server architecture you
(23:30):
know.
Basically it's not difficult tothink well, wait a second,
isn't there a bottleneck insidethe machine?
And this must require a reallysuper hot, rotted machine, you
know.
So maybe the cost savings isoffset by super expensive
hardware, and I think it'simportant to note that the
(23:53):
reason why this is possible is,first of all, the VPU is built
on NVMe architecture, so it'susing the exact same storage
protocol as your hard drive, youknow, as the SSDs that are in
the machine.
And what we have done, whatNetEnt has done, is actually
(24:17):
created a peer-to-peer sharinginside the DMA.
So, basically, the GPU willoutput a frame, a rendered frame
, and it's transferred literallyinside memory so that then the
VPU can pick that up, encode itand there's effectively zero
(24:39):
latency, at least in terms ofthe latency is so low because
it's happening in the memorybuffer.
And so if anybody's listeningand raising an eyebrow,
wondering, well, wait a second,you know, surely there's a
bottleneck, and especially ifyou're talking 60 frames per
second which, by the way, ourbenchmarks are generally always
(25:01):
at 60 frames per second because,you know, unless it's real
casual games, you need thatframe rate to really deliver a
great experience, even aboveresolution, in some cases it's
better to get the frame rate upthan to increase the size of the
frame.
Absolutely, absolutely,absolutely.
(25:21):
Yeah, let me just pause here andsay that we would love to have
questions, and so feel free, onwhatever platform you're if
you're on YouTube or LinkedIn orwherever you're watching us
right now just type in and Iwill try and pick those up.
(25:42):
I have looks like we alreadyhave one, so you know why don't
we?
I think this is actually areally good one.
I'm gonna pick this up righthere, but feel free to enter
questions in the chat.
So, Olivier, the question is Ilive in a country where stable
(26:08):
internet is not always availableand, by the way, I would say
that you know this isn't only,you know, a country issue.
Internet varies, right, and theexpectation of users is more
and more that.
You know they don't think aboutthe fact that I'm in a car.
You know I happen to be in anarea where there's great
(26:29):
coverage, but you know, sevenmiles down the road that changes
right.
They want to keep playing andkeep enjoying this great
experience.
So the question is I live in acountry where stable internet is
not always available.
How will this affect the gamingexperience?
And you know, yeah, I mean,that's the question.
(26:52):
So, what's your experience andhow are you guys solving for
this?
Speaker 2 (26:56):
You see, in Netflix or Spotify, right, you can
actually buffer content so that,even if your bandwidth is, you
know, a bit clumsy, you canactually store that content in a
CDM and keep the experience,you know good enough, right, or
you can download the thatcontent in a CDN and keep the
experience good enough, or youcan download the video and make
it work.
So, definitely you have someway to solve that problem in, I
(27:19):
would say, cold media, mediathat you can encode in one way,
then stream later.
In games, this is completelydifferent.
Yeah, you can't do that Becausewe have to encode, stream,
deliver and then you take theiteration right away.
So if your bandwidth is notenough, right.
If the quality of the bandwidthis not enough, and not only in
(27:41):
terms of the size of thebandwidth but also in terms of
characteristics, the latency,how this latency is stable, and
so on, then the experience willbe great.
So what we've been doingactually with Ericsson is to use
5G networks and to definespecific characteristics of what
(28:01):
is a slice in the 5G network,so we can tune the 5G network to
make it fit for gaming and tooptimize, basically, the
delivery of gaming with 5G.
So we think that 5G is going toget much faster in those
regions where actually theinternet is not so great.
We've been deploying theBlacknet service in Thailand, in
(28:27):
Singapore, in Malaysia, now inthe Philippines and so on, and
this has allowed us to actuallyreach people in regions where
there is no cable or bandwidthwith fiber and this kind of
thing.
So, look, I'm not going tosolve the problem where
(28:47):
bandwidth is not available, butmaybe bandwidth will come faster
with 5G and that could be thesolution.
Speaker 1 (28:54):
Yeah, I want to make a comment there and thank you
for the answer we are seeing.
So it's very interesting andI'll use India as an example.
So for years in video streaming, the Indian market was used as
an example of where it was verydifficult to deliver high
(29:19):
quality and especially if youwanted to deliver, say, 720p and
1080p was almost assumed at acertain period of time like it's
not even possible because thenetwork capacity and the speeds
were just so low.
What has happened is and Indiais a great case study here, but
(29:43):
it's really almost all regionsof the world as these
infrastructures, these wirelessinfrastructures, have been
upgraded, they leapfroggedliterally from 3G or you know in
some cases even you know 2.5Gand you know before and just
went all the way to 5G.
(30:03):
And so in the last five yearsthere has been such a
fundamental shift in bandwidthavailability that in some cases,
some of these regions of theworld, not only is it definitely
no longer true that they'reslow, they're faster than some
(30:23):
of the more developed countries.
So I do want to make thatstatement there.
One question, olivier, can youtalk about?
Is this WebRTC?
What protocols you're using?
There's a lot of talk right nowabout QUIC and I think that
(30:45):
would be interesting for some ofthe listeners who might be
wondering even what protocolsyou're using.
Speaker 2 (30:51):
So we use standard codecs to start with the bottom
line.
We have not invented codecs.
We have been in thestandardization industry of
audio and video for quite someyears and I think you have great
experts here doing greattechnology.
And this technology is actuallyembedded into the chipset, into
the hardware, so actually youcan rely on hardware encoding
(31:15):
and decoding capabilities.
So we do think standard codecsis basically a must-have, right?
Of course, you need toconfigure them the right way,
because you have to codereal-time, okay, so you cannot
use particular techniques.
You know to wait for a coupleof frames to compute more, so
you have to optimize this, butbasically we use standard
(31:35):
collects.
Then, on the protocols, on topof this, we have actually a
large variety of protocol okay,we depend on the device on which
you are streaming.
So it goes from fullproprietary protocol that we
have invented and patented inBlackNut to standard WebRTC.
So, if you look at devices likeSamsung and LG, which are
(32:00):
basically the top manufacturers,I think the service has been
launched on LG we are going toannounce, I think, our launch
with Samsung in a very shorttime and these devices support
WebRTC, and that's basically theonly way to implement and to
support the cloud gamingsolution efficiently.
(32:22):
So short answer we use a widerange of protocols, always the
one that is the most appropriateand provides the best
experience to the end-user.
We are using that, of coursenew protocol, new standards,
experimenting this but I wouldsay for the main streamline new
solution, we use our ownsolution, plus WebRTC, when this
(32:45):
is the only standard that therethe end-to-end latency targets.
Speaker 1 (32:50):
I think previously you made the comment about you
know like 80 milliseconds, butgive us some guidelines.
What is obviously the answer isas low as possible, but what's
the upper limit where the gameexperience just falls apart?
You know, it's just notplayable.
Speaker 2 (33:09):
You know that the limit for conversional video is
about 150 milliseconds.
For playing games this is muchlower, probably half of it.
So I think you know you can geta reasonably good experience at
80 milliseconds, or actuallymost of the games.
That does not require you knowthis kind of fast reaction.
But then if you want to go toFPS or this kind of thing that
(33:32):
really need to nearly bereactive at the frame accuracy,
which is very, of course,difficult in cloud gaming, you
need to go down to the 30milliseconds and lower.
And that I think is onlyfeasible if you have the network
that allows for it, becauseit's not only about the encoding
part, the server side, and theclient side, it's also on where
(33:57):
the packets are going throughthe networks, because you can
have the most efficient systemsin terms of encoding latency and
decoding latency.
But if you pocket, instead ofgoing directly from the server
to the end user, go here andthere and transit in many places
, your experience will be crappy.
And you know, Mark, this isactually a real issue because we
(34:18):
, for example, had a greatdemonstration with Ericsson in
Madrid, you know, in Barcelona,right for the Mobile World
Congress, and we had servers inMadrid for the Mobile World
Congress, and we had servers inMadrid.
But when we first made thefirst test, we discovered that
the packet was going from Madridto Paris and back to Barcelona,
right, okay.
So we needed a bit ofintelligence and technology to
(34:41):
make this connection asefficient as possible.
Speaker 1 (34:44):
Tell us about Black Nut.
What exactly do you guysdeliver?
Speaker 2 (34:50):
We provide basically a cloud gaming service which is
let's categorize it as a game asa service.
This means that for thesubscription fee per month, you
get access to the real stuff.
You get access to 700 games.
We are adding 10 to 15 newgames per month, which is, I
think, the fastest pace in termsof increasing games on the
(35:12):
market, and we provide thisexperience on all single devices
that can actually receive avideo.
So that's what we do and wedistribute this service either
B2C, so direct to the consumer.
So if you go on your BlackNetwebpage, you know you can
subscribe, you can access to thegames.
But we also distributed throughcarriers, so telecommunication
(35:36):
carriers, operators all over theworld.
We currently have about 20signed agreements with the
carriers live actually more than40 signed and we are signing
and delivering one to two newcarriers per month, you know.
So that basically is 40 signand we are signing and
delivering one to two newcarriers per month.
So that's basically the pacewhere we are in BACnet and the
(35:56):
choice to use carriers here isfor the reason I explained to
you, that it's good to havegreat service.
Speaker 1 (36:01):
Optimization of the network right.
Speaker 2 (36:05):
You need to know where the packets are going.
You need to make sure thatthere is some form of CDN for
cloud gaming that is in placehere that makes the experience
optimal.
Speaker 1 (36:14):
Yeah, it completely makes sense to me, especially
because you mentioned the 5Goptimization.
And obviously, carriers.
They've been investing now foryears in building out their 5G
networks but they're alwayslooking for reasons to drive
more value and to really extractthe full potential off the 5G
(36:36):
or out of the 5G investment.
So, yeah, it really makes sense.
Speaker 2 (36:41):
That's the kind of thing we are doing as well with,
you know, our partner,radianark, and we are putting
the server at the edge of thenetwork.
So inside you know thecarrier's infrastructure, so
that you know the latency isreally super optimized.
So that's one thing that is keyfor the service.
Speaker 1 (36:57):
What is the architecture of that edge server
?
What's in it?
What CPU, GPU, VPU, like?
Describe that.
Speaker 2 (37:07):
We started with a standard architecture, you know,
with CPU and GPU, and now, youknow, with the current you know
VPU architecture, we are puttingactually a whole server
consisting in AMD GPU, you knowNetIn GPU, and basically we
build the whole package so thatwe put this in the
(37:27):
infrastructure of the carrierand we can deploy the BlackBerry
Cloud Gaming on top of it.
Speaker 1 (37:33):
And are you delivering to only a handful of
fixed resolutions, or like if Iwas on a TV, for example, do I
get 4K, or do you limit to 1080p, or how do you handle that?
Again, great question.
Speaker 2 (37:52):
We actually can handle multiple resolution,
right.
I think we can stream from 720pup to 4k.
Uh, the technology basicallyhas no limits for it, right?
Um, and streaming you know 4kor even you know 8K is a problem
that has somehow been solvedalready.
From a technical matter.
The question is again the costand the experience.
(38:16):
Streaming 4K on a mobile devicedoes not really make sense.
I think the screen is a bitmore, so you can screen a
smaller resolution and that'ssufficient.
On a TV, likely you need tohave a bigger resolution, even
if actually there is a greatupscale available on most of the
(38:38):
TV sets.
We stream 720p on Samsungdevices and that's super great,
right, but of course, scanningup to 1080p will provide a much
better experience.
So, on TVs and for the gamethat require it, I think we are
indeed streaming the serviceabout 1080p for the game that
(39:00):
requires it.
Speaker 1 (39:01):
Do you also find that frame rate is almost more
important than resolution Forcertain?
Speaker 2 (39:08):
games, absolutely, but you know again, it is game
dependent.
It's game.
Yeah, if you are on FPS, youprobably.
If you have the choice, okay,and you cannot stream 1080p, you
would probably stream 720p at60 FPS rather than 1080p at 30
FPS, right?
Yes, yes, if you have to makesome trade-off, but if you have
(39:30):
different games where thetextures, the resolution is more
important, then maybe you willactually select more 1080p and
30fps resolution and thetechnology we build is actually
fully adaptable.
Ultimately, you should notforget that there is a network
in between and even if,technically, you can stream 4K
(39:51):
or 8K, the networks may notsustain it and then actually you
will have less good experiencestreaming 4K than actually a
1080p 60fps resolution.
Speaker 1 (40:04):
Okay, I see, a question just came in
(40:28):
no-transcript.
You actually talked about howmany publishers you have.
You did talk about every monthyou're onboarding, you know, I
think, 10 or 12 new games, butyeah, so are there geographical
restrictions?
How can someone access this?
Speaker 2 (40:48):
great, let's start's start with content.
Okay, indeed, we have, you know, more than several games right
now, 10 to 15 new games permonth, and we actually try not
to have geographical limitationon the content.
Okay, so it's been.
The content we have on thecatalog is, from a licensing
point of view, availableworldwide.
(41:10):
That's basically what we do.
We do have exceptions, as usual, but basically a large part of
the catalog is availableworldwide.
Now, deploying this catalog ondifferent region, we are
available in more than 45countries.
We definitely need to haveservers that are close enough to
(41:32):
the end user so that thestreaming experience is good
enough, and we think that aradius of between 750 to 1,500
kilometers probably is themaximum kilometers, you know,
probably the maximum.
So I think we will actually putsome point of presence, you
(41:57):
know, in those geographicalareas so that, basically, the
latency, limited by the speed oflight right, does not, you know
, harm the service.
So, of course, you know, if youlook at it, we have europe very
much, you know, covered.
We have us and canada very muchcovered.
We have a large portion ofsoutheast asia, korean and japan
(42:18):
very much covered.
We are now expanding in latinamerica, which is a bit harder.
We are strong presence now aswell in in the middle with
partners like STC in the regionand of course we have some zones
that are less covered.
Africa is not well covered atall.
(42:39):
South Africa is, but basicallythe rest of Africa is a bit
harder to reach.
Speaker 1 (42:45):
By the way, what is the website?
Why don't you give out the?
Speaker 2 (42:50):
URL there wwwblacknutcom.
Speaker 1 (42:54):
Blacknutcom, all right.
Speaker 2 (42:57):
I think, try the service.
We'd be very happy to supportand provide feedback.
Speaker 1 (43:05):
It's super exciting and you know, as I, as I said in
the in the beginning, for mepersonally, having been, you
know, really, in the very earlystages of of the transition from
physical entertainment deliveryI'm talking about movies, you
know, specifically like DVDs tostreaming.
(43:26):
I'm just super excited to alsonow, 15 years later, be there
with games.
And there's a lot of work to bedone and, as you pointed out,
you know, the experience isabsolutely not exactly mapped.
We can't replace, you know,throw out the console yet.
(43:47):
You know, throw out the consoleyet, uh, and but the, the
opportunity to bring, you know,really, the gaming experience to
, um, to a much wider audience,is really enabled, you know,
with streaming.
So, um, by by the way, uh.
So I think there's a follow-onquestion here Do you have
(44:08):
infrastructure in South Africayou mentioned?
Africa is not covered as well.
Speaker 2 (44:13):
Yes, we do have the capacity to deploy the service
in South Africa.
Speaker 1 (44:18):
To deploy in South Africa.
Okay, great, great.
Well, we're right up againsttime and thank you for everyone
who joined us.
Live Really appreciate it.
And thank you, Olivier, it'samazing what you built and we're
super excited to be workingwith Black Nut.
Thank you everyone.
Speaker 2 (44:36):
Thanks, Mark.
This episode of Voices of Videois brought to you by NetInt
Technologies.
If you are looking for cuttingedge video encoding solutions,
check out NetInt's products atnetintcom.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!