• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Next-Gen PS5 & XSX |OT| Console tEch threaD

Status
Not open for further replies.

xool

Member
Yeah i am lowering back down to team 399 and 9tfs. People forget that Jim Ryan Sony ceo says that they want a faster transition this upcoming gen. Selling a ps5 for 499 doesn’t match that goal
Did read my whole post on how Jim Ryan wants a faster transition

Not what he said though - he was talking about backwards compatibility (twinfinite.net/2019/05/sony-jim-ryan ) :

We will leverage Backwards Compatability to transistion our community to Next Gen faster and more seamlessly than ever before

[..]

[PS4] Will remain the engine of engagement and profitablity for the next three years

 
Last edited:

CrustyBritches

Gold Member
Doesnt make sense considering 64CU @1500Mhz would consume the same or less
That sounds expensive. It's cheaper to have a smaller chip because of better yields. We'll have to see how the Navi 10 Pro AiBs do at higher clocks. Maybe it's just a fast chip and won't hit the wall as fast as old GCN cards? Think of it like a modern day GTX 1060...most could hit 2GHz easy breezy and still pull under 140W compared to what happens to RX 480/580...
 
I have a few more questions.

1) DF next gen xbox scarlett/ps5 spec analysis video stated that the next gen video game GPU will start off "roughly equivalent to 1080TI which is not bad" WTF????!!!! That's really lowballing it!
2) If scarlett and ps5 are gonna be released in 2020, why wont it have Wifi 6?!!
The 1080TI is no slouch. You’re not going to get a console up in the 2080ti range unless you want to pay over $1000.
 

SonGoku

Member
That sounds expensive. It's cheaper to have a smaller chip because of better yields. We'll have to see how the Navi 10 Pro AiBs do at higher clocks. Maybe it's just a fast chip and won't hit the wall as fast as old GCN cards? Think of it like a modern day GTX 1060...most could hit 2GHz easy breezy and still pull under 140W compared to what happens to RX 480/580...
An RTX2060 is 445mm2 and costs $349
A 380-400 mm2 chip will go through more costs reductions compared to a smaller power hungry chip, making it cheaper long term

If 36CUs at 1.8ghz pull 140W there's even more reason to go with a 64CU chip for maximum performance while remaining cost effective and power efficient.
 
Last edited:

SonGoku

Member
But PS5 die size is 316mm² so this is all in vain..
giphy.gif
 

FrostyJ93

Member
So looks like a new Ape Escape may be coming. So looks like that RuthenicCookie guy on reddit definitely has inside knowledge. The only thing he got wrong was anthem delay.
 

CrustyBritches

Gold Member
If 36CUs at 1.8ghz pull 140W there's even more reason to go with a 64CU chip for maximum performance while remaining cost effective and power efficient.
I didn't mean that it will pull 140W, that's just an example of a chip comfortable at 2GHz OC without much of an impact on power consumption. Most RX 480's maxed at 1390MHz, but looking at RX 580 you have 1.5GHz with over 200W consumption.

With a smaller chip it has better yields than a bigger chip, so more working chips for the same wafer.
 

R600

Banned
Remember, best leaks last gen came from completely obscure sources. Not Matt, not IGN, not GI - sweetvar26 from GAF who got friend working at AMD.

I am 100% sure Gonzalo chip, that is undoubtedly console one (G in codename) and has ID associated with Ariel (PS5), had engineering sample back in January, and QS in April. After that, May 21, Reddit leak comes from guy somewhere in Asia, who tells us technical details of PS5 dev kit that has just passed OQA. Next day, his account is deleted.
 
Last edited:
People still haven't figured out how "leaks" work it seems.

Items in leaks that are corroborated by future events are things that don't matter to strategy.

They are given as a freebie to lend weight to the facts that you want to distort.
 

SonGoku

Member
I didn't mean that it will pull 140W
I know lol but if it did it makes the case for a bigger chip
With a smaller chip it has better yields than a bigger chip, so more working chips for the same wafer.
8 disabled CUs will make up for that and as yields improve costs go down vs a small power hungry chip where there's less room for cost reductions.
Any way you look at it the bigger chip delivers more power efficiency and costs reductions
Remember, best leaks last gen came from completely obscure sources. Not Matt, not IGN, not GI - sweetvar26 from GAF who got friend working at AMD.

I am 100% sure Gonzalo chip, that is undoubtedly console one (G in codename) and has ID associated with Ariel (PS5), had engineering sample back in January, and QS in April. After that, Reddit leak comes from guy somewhere in Asia, who tells us technical details of PS5 dev kit that has just passed OQA. Next day, his account is deleted.
You are free to believe what you please, just dont be surprised people dont take your obscure source as gospel, those are a dime a dozen on pastebin and reddit and many reddit leakers delete their accounts
 
Last edited:

R600

Banned
People still haven't figured out how "leaks" work it seems.

Items in leaks that are corroborated by future events are things that don't matter to strategy.

They are given as a freebie to lend weight to the facts that you want to distort.
They can also easily leak when things go into production, such as next gen chips. This is how Apple stuff leaks all the time. This is how I believe PS5 dev kit leaked, someone from QA leaked whats inside the box because no one can open dev kit once its sent to studio. Only Sony people and people who are actually on production line can identify whats on motherboard.

I am 100% PS5 dev kit leaked.
 
Last edited:
They can also easily leak when things go into production, such as next gen chips. This is how Apple stuff leaks all the time. This is how I believe PS5 dev kit leaked, someone from QA leaked whats inside the box because no one can open dev kit once its sent to studio. Only Sony people and people who are actually on production line can identify whats on motherboard.

I am 100% PS5 dev kit leaked.

And that can't be done as a controlled "leak" because?

Did it turn out to be a money shot or another tease? ;)
 

LordOfChaos

Member
With 1.84 Tflop only :messenger_sunglasses:

giphy.gif


Just finishing that, and 100%, can't wait to see what those guys can do with RDNA on anything north of 9TF and the new higher memory cap and 4x the CPU performance and 8X the AVX performance. Would the higher end estimates be nice, sure, but we'll get what we get and Sony devs will put it to amazing use.

You know, my number one (and really only) complaint with the game is something Sony and MS seem keen to solve, it's all the crap it did to hide load times (go back to the gateway, use the portal to shift realms, repeat, etc). That's a known quantity that's being taken care of in next gen and I'm pumped for that.

Well, other than Sigrun, FUCK Sigrun!
 
Last edited:

R600

Banned
Devkit leak doesnt give priority to one leak over the other
Just to give you a counter example: Whats to say Benjii 12.9TF leak isnt from a Navi chip?
Because Benjis info is from December 2012, when SOC containing Navi and Zen2 couldnt have been tapped out. Additionally, it absolutely doesnt fit timeline compared to PS4, where first real SOC dev kits get into hands of 3rd party in late 2012, year before console was retail.

If they had 12.9TF SOCs by late 2012, there would be no need to sit on it for 2 years before releasing new console, as that chip is as good as done.

Timeline goes like this :

1. Gonzalo engineering sample appears in January. CPU base clock - 1600MHZ containing Ariel ID and GPU clocked at 1GHZ

If you follow AMD you will notice ES having seriously low clock speeds

2. Gonzalo APU re appears, this time as QS (quality sample) in April. CPU is 8 core running 3.2GHZ and GPU running 1.8GHZ

3. Sony officially announces PS5 where Cerny mentions that current Dev Kits are "slow", duo to SSD tech having alot of propriatary parts and current dev kits not having it (essentially, PCs?

4. Reddit post by guy, cleary from Asia (as he "celebrated" "Cake Day" on 21st, while post was made on 20th of May GMT) leaks alleged PS5 dev kit from OQA. Memory types, die size, SSD and VRM controllers. His avatar from Reddit is electric circuit. Deletes username 2 days later.

5. What is Gonzalo? If Gonzalo boost is 1.8GHZ, and 5700 that uses 180W max is 1.72GHZ max boost, how can this APU be big one? It makes 0 sense. Either its a big one, with lower clocks, or its smaller one with higher clocks.

I want to add that NOTHING we know of Navi chip - die size, TDP, CU arrangement etc. tells us to expect chip with 50+ CUs and RT cores, especially if clocked that high. This is repeat of 2012 and time when we expected 3+ TF GPU machines (as 7970 had 3.8TF) and got 1.2TF and 1.8TF ones.
 
Last edited:

SonGoku

Member
If they had 12.9TF SOCs by late 2012, there wouls be no need to sit on it for 2 years before releasing new console.
You mean late 2018 right?
A chip tape out design can happen 1-2 years in advanced, that doesn't make it viable for mass production: low yields and extensive testing required
I want to add that NOTHING we know of Navi chip - die size, TDP, CU arrangement etc. tells us to expect chip with 50+ CUs and RT cores
Calculations point towards a 64CU-72CU GPU fitting on a 380-400 mm2 die RT sillicon included

Lastly a 400mm2 die on 7nm would translate to ~320 mm2 on 7nm EUV. That would mean 56-64CUs (enabled) at 1.8GHz. Goddamn
 
Last edited:

Imtjnotu

Member
I know lol but if it did it makes the case for a bigger chip

8 disabled CUs will make up for that and as yields improve costs go down vs a small power hungry chip where there's less room for cost reductions.
Any way you look at it the bigger chip delivers more power efficiency and costs reductions

You are free to believe what you please, just dont be surprised people dont take your obscure source as gospel, those are a dime a dozen on pastebin and reddit and many reddit leakers delete their accounts
is kugoson still in the lead for leaks?
 

R600

Banned
You mean late 2018 right?
A chip tape out design can happen 1-2 years in advanced, that doesn't make it viable for mass production: low yields and extensive testing required
They can, in extreme situations, not console business where everything runs super tight. AMD actually told us in January 2018, earnings call, that Zen2 wil be taped out by the end of 2018 Back then, Navi was in huge problems mind you and was rutinely ingnored by AMD.


So, by end of 2018 AMD didnt even tape out Zen2, nor Navi, so how would Sony tape out 12.9TF Navi/Zen2 based SOC two years before release? They didnt.

Calculations point towards a 64CU-72CU GPU fitting on a 380-400 mm2 die RT sillicon included
No calculations say that. Calculations say that Xbox One X SOC has 44CUs on 359mm² die size with PC counterpart that has 36CUs on 231mm². In addition to this, Zen2 will take more space on die then Jaguar, and of course hardware acc RT. So how do you fit 50% more CUs + RT hardware in console box when PC part is 250mm² with TDP of 225W?
 
Last edited:

SonGoku

Member
They can, in extreme situations, not console business where everything runs super tight. AMD actually told us in January 2018, earnings call, that Zen2 wil be taped out by the end of 2012. Back then, Navi was in huge problems mind you and was rutinely ingnored by AMD.
Fix your quote
So, by end of 2018 AMD didnt even tape out Zen2, nor Navi, so how would Sony tape out 12.9TF Navi/Zen2 based SOC two years before release? They didnt.
How do you know this, source? I find it very hard to believe a chip launching this year didnt tapeout on 2018
To answer your question, AMD gives great importance to its semi custom bussinness. Its not out of the real of posibility for them to tape out console chips earlier than their own lineup to accommodate for the extensive testing required , PS5 chip being the first navi tapeout falls in line with early AMD made navi for Sony rumors
So how do you fit 50% more CUs + RT hardware in console box when PC part is 250mm² with TDP of 225W?
Proelite calculations fit 64CUs on a 385mm2 APU RT hw included. Using his math 72CUs fit on 400mm2
tdp is lower due to lower clocks and voltage and better yields
 
Last edited:

R600

Banned
Fix your quote

How do you know this, source? I find it very hard to believe a chip launching this year didnt tapeout on 2018
To answer your question, AMD gives great importance to its semi custom bussinness. Its not out of the real of posibility for them to tape out console chips earlier than their own lineup to accommodate for the extensive testing required , PS5 chip being the first navi tapeout falls in line with early AMD made navi for Sony rumors
They can give all importance in the world to their semi custom business, but there is a plam that needs to be followed. Sony knew roadmap as did MS, they never tapped out chips 2 years before console release, that never happened. Zen2 wasnt tapped out either by AMD, nor was Navi, so why would be Sonys SOC 2yrs before release?

Proelite calculations fit 64CUs on a 385mm2 APU RT hw included. Using his math 72CUs fit on 400mm2
tdp is lower due to lower clocks and voltage and better yields
Nah, it just doesnt fit, its right there in front of you. PC part with 40CUs on same process is 251mm². PC part with 36CUs that was found in Scorpio had 231mm². How much did Scorpio with 8CU more take up on die? 360mm², with measly Jaguar. So where do you fit even bigger chip when MS maxed out 44CUs on 360mm² die, no RT and smaller CPU?

This is not even talking TDP that is through the roof with 40CU part. 225W. 40W more then Pitcairn (GPU that ended up underclocked and with 2 CUs disabled in PS4) and 40 more then the PC counterpart of one found in PS4Pro.
 
In addition to this, Zen2 will take more space on die then Jaguar
Octa-core Zen at 7nm will require the same amount of die space as octa-core Jaguar did at 28nm. There's nothing fancy about it.

Some detailed calculations: https://www.resetera.com/threads/se...read-summary-in-op.33810/page-48#post-6490656

At worst, maybe they'll axe half of the L3 cache to improve yields, assuming it doesn't effect gaming performance that much.

Either way, we know for a fact that 7nm offers 3.3x density improvement compared to 16nm FinFET. One of the biggest leaps ever.
 
Last edited:

R600

Banned
Octa-core Zen at 7nm will require the same amount of die space as octa-Jaguar did at 28nm. There's nothing fancy about it.

Some detailed calculations: https://www.resetera.com/threads/se...read-summary-in-op.33810/page-48#post-6490656

At worst, maybe they'll axe half of the L3 cache to improve yields, assuming it doesn't effect gaming performance that much.

Either way, we know for a fact that 7nm offers 3.3x density improvement compared to 16nm FinFET. One of the biggest leaps ever.
Ok that is good, didnt know Jaguar took so much space. Still, that TDP, I dont know where you guys find belief in that. 225W is 9.75TF Navi, that spells game over for anything over 10TF. Maybe they csn get more CUs at lower clocks, but I seriously doubt they are going for it. There is a reason they have been on mute regarding TF number.
 

SonGoku

Member
they never tapped out chips 2 years before console release, that never happened.
uhm yes that's standard procedure
Zen2 wasnt tapped out either by AMD
Source?
why would be Sonys SOC 2yrs before release?
To accommodate extensive testing required
Gonzolo early this year isnt much different from late 2018, cherry picking leaks?
Nah, it just doesnt fit, its right there in front of you. PC part with 40CUs on same process is 251mm². PC part with 36CUs that was found in Scorpio had 231mm². How much did Scorpio with 8CU more take up on die? 360mm², with measly Jaguar. So where do you fit even bigger chip when MS maxed out 44CUs on 360mm² die, no RT and smaller CPU?
Yes it does.
5700:
GDDR6 phy controller: 4.5mm x 8
Dual CU: 3.37mm x 20
4 ROP cluster: .55mm x 16
L1+L2+ACE+Gemotry processor+empty buffer spaces + etc: 139mm

378 mm2 estimate
75mm for CPU
45mm for 10 GDDR6 controllers
8.8mm for ROPs
140mm for buses, caches, ACE, geometry processors, shape etc. I might be over estimating this part as the 5700 seems to have lots of "empty" areas.

We have ~110mm left for CUs + RT hardware. There is enough there for ~30 dual CUs and RT extensions

He is even lowballin CUs using 5700 empty spaces and adds up to 378.8 mm2

Using his estimates and maintaining the empty spaces on 7nm
80CU APU = 412.6 mm2
72CU APU = 399.1 mm2
64CU APU = 385.6 mm2

BTW CUs take less than 50% of GPU die, its not a linear scale
 
Last edited:
Ok that is good, didnt know Jaguar took so much space. Still, that TDP, I dont know where you guys find belief in that. 225W is 9.75TF Navi, that spells game over for anything over 10TF. Maybe they csn get more CUs at lower clocks, but I seriously doubt they are going for it. There is a reason they have been on mute regarding TF number.
My take on that is that 7nm HPC is an immature process (low yields, high power consumption/voltage) right now.

Have you wondered why Nvidia is waiting for 2020 to manufacture and release their 7nm Ampere GPUs? They must know something that we don't, right?

AMD wanted the "7nm first!" bragging rights (sounds dumb, but it's true). It clearly didn't work that well for GPUs (Radeon VII power consumption and Navi 10 being too expensive for a 250mm2 mid-range GPU die). Zen 2 CPUs will fare better because of their chiplet structure (ask yourself why they did that, instead of going monolithic).

I'm hopeful that by mid to late 2020 this process will have improved immensely for big monolithic chips.
 

SonGoku

Member
225W is 9.75TF Navi, that spells game over for anything over 10TF. Maybe they csn get more CUs at lower clocks, but I seriously doubt they are going for it.
Its all about voltage and yields
The voltage necessary to maintain 1950Mhz stable is responsible for that, AMD is pushing the card beyond perf/watt sweet spot for maximum performance
As yields improve the voltage required to sustain the same clocks decreases
There is a reason they have been on mute regarding TF number.
Yes there is, Sony hasn't formally anounced PS5 yet (same for ms), didn't even mention ram.
Makes no sense to give hard numbers this far in advanced before the big reveal.
 
Last edited:

DanielsM

Banned
Ok that is good, didnt know Jaguar took so much space. Still, that TDP, I dont know where you guys find belief in that. 225W is 9.75TF Navi, that spells game over for anything over 10TF. Maybe they csn get more CUs at lower clocks, but I seriously doubt they are going for it. There is a reason they have been on mute regarding TF number.

It makes absolutely zero sense to be talking hardware 12-17 months out, generally speaking, it kills current hardware sales. I have no idea why Microsoft officially announced Xbox Next without really showing hardware yet again this far out. (what not to do)
 
Last edited:

Imtjnotu

Member
Its all about voltage and yields
The voltage necessary to maintain 1950Mhz stable is responsible for that, AMD is pushing the card beyond perf/watt sweet spot for maximum performance
As yields improve the voltage required to sustain the same clocks decreases

Yes there is, Sony hasn't formally anounced PS5 yet (same for ms), didn't even mention ram.
Makes no sense to give hard numbers this far in advanced before the big reveal.
with ps4 they changed ram numbers right up to the Reveal in Feb. all leaks and i mean all, for orbis, were 4gb of GDDR5
 

LordOfChaos

Member
Zen 2 was supposed to tape out in 2H 2017, at least

"Our goal is to be very competitive in terms of our long-term roadmap. If you look at the foundry’s 7nm roadmap compared to some of the other technologies out there it’s actually really competitive. I think the gap between the foundry roadmap and the Intel roadmap has gotten a lot closer. Our goal is to be aggressive with 7nm technology. We will be doing tape outs later this year and as we get closer to production will give more insights there. But the idea is to be more competitive throughout the portfolio." -SuBae



 
Last edited:

xool

Member
Digital foundry said 1080 performance in their video. I don't think they said Ti.
I just watched the recent one and didn't catch that - they suggested somewhere between Vega56 and 64 but without actually betting on it ..



Also from the video :

  • Scarlet die size 365-380mm2 (Beyond3d estimate)
  • Zen2 8 core CPU 70-80mm2
  • AMD 5700 (XT) 251mm2 (40CU)
  • Traditional console die size ~320mm2
Puts low expectations at equivalent to 8 core Zen2 + 5700 GPU, and based on the scarlet die size guess that estimate gets a boost of ~50mm2 extra, presumable for more CUs ..

Supposedly the 5700XT equals/beats 2070, and the 2070 wins/loses/equals the 1080ti depending on bench .. so with a little extra cheese the theoretical next gen scarlett should equal exceed a 1080ti.

I hope. Not too shoddy? (Where expectations higher?)
 

CyberPanda

Banned
I just watched the recent one and didn't catch that - they suggested somewhere between Vega56 and 64 but without actually betting on it ..



Also from the video :

  • Scarlet die size 365-380mm2 (Beyond3d estimate)
  • Zen2 8 core CPU 70-80mm2
  • AMD 5700 (XT) 251mm2 (40CU)
  • Traditional console die size ~320mm2
Puts low expectations at equivalent to 8 core Zen2 + 5700 GPU, and based on the scarlet die size guess that estimate gets a boost of ~50mm2 extra, presumable for more CUs ..

Supposedly the 5700XT equals/beats 2070, and the 2070 wins/loses/equals the 1080ti depending on bench .. so with a little extra cheese the theoretical next gen scarlett should equal exceed a 1080ti.

I hope. Not too shoddy? (Where expectations higher?)

Check my post above. They say it between 17:15-17:31
 

xool

Member
Go about 17:15-17:31 in.
k thnx . I was looking at the more recent one

I think most people would be deeply disappointed with next gen if it didn't meet 1080 performance levels - 1080 has been the high-performance mainstay comparison for so long I'd want to at least meet that next gen
 

JohnnyFootball

GerAlt-Right. Ciriously.
PS4s dominance this gen would have probably been considered impossible if someone predicted it in 2012-2013 on the heels of an incredibly successful 360 run.
I’d agree with this statement if your dates were 2010-2011. By 2012, the tide was turning in Sony’s favor and by then the Xbox was going downhill and we’re going all in on Kinect.
 

bitbydeath

Gold Member
Sony have lodged a patent to further improve load times by automatically splitting segments for all games. (Which to me sounds like what they already do for first party but instead now have the ability to enforce it for all titles.)


Edit: Where’s my link preview gone? :-(
 
Last edited:
Sony have lodged a patent to further improve load times by automatically splitting segments for all games. (Which to me sounds like what they already do for first party but instead now have the ability to enforce it for all titles.)


Edit: Where’s my link preview gone? :-(
Please for the love god just fucking release the ps5 already lmao. Give GOW 2 with this shit plzzzzzzzzzz
 
I want to add that NOTHING we know of Navi chip - die size, TDP, CU arrangement etc. tells us to expect chip with 50+ CUs and RT cores, especially if clocked that high. This is repeat of 2012 and time when we expected 3+ TF GPU machines (as 7970 had 3.8TF) and got 1.2TF and 1.8TF ones.
If MS had practiced the Scorpio paradigm back in 2013 (huge die, GDDR5, focused on gaming usage, no eSRAM/TV/Kinect) we would have easily gotten a 3TF console at 28nm for $499.

Instead we got gimped consoles due to having a price target of $399 (XB1 cost the same as the PS4 if you remove the Kinect shenanigans).

Do you reckon 2020 will be a repeat of the 2013 situation? Judging by Cerny's comments indicating a premium machine/price point, I think it's highly unlikely.

Of course Sony can get complacent if they feel like nothing threatens them, instead of taking advantage of their golden goose to strengthen their stronghold:

DbykEZ1WAAAX5JA.png


I'm betting my ass off that Fortnite royalties have been funding PS5 R&D since 2017. :)

Regarding power:


top-factor-driving-ps4-purchases-is-better-resolution-survey-142494438801.png


Power matters during the first 1-2 years (not a lot of must-have/exclusive games). Right now the PS4 has a far bigger library than it had back in 2013-2014, so it's mainly games that fuel the momentum right now. It wasn't like that in the beginning of the generation. That's the difference.
 
Last edited:

StreetsofBeige

Gold Member
If MS had practiced the Scorpio paradigm back in 2013 (huge die, GDDR5, focused on gaming usage, no eSRAM/TV/Kinect) we would have easily gotten a 3TF console at 28nm for $499.

Instead we got gimped consoles due to having a price target of $399 (XB1 cost the same as the PS4 if you remove the Kinect shenanigans).

Do you reckon 2020 will be a repeat of the 2013 situation? Judging by Cerny's comments indicating a premium machine/price point, I think it's highly unlikely.

Of course Sony can get complacent if they feel like nothing threatens them, instead of taking advantage of their golden goose to strengthen their stronghold:

DbykEZ1WAAAX5JA.png


I'm betting my ass off that Fortnite royalties have been funding PS5 R&D since 2017. :)

Regarding power:


top-factor-driving-ps4-purchases-is-better-resolution-survey-142494438801.png


Power matters during the first 1-2 years (not a lot of must-have/exclusive games). Right now the PS4 has a far bigger library than it had back in 2013-2014, so it's mainly games that fuel the momentum right now. It wasn't like that in the beginning of the generation. That's the difference.
Xbox X was $500 at 6TF. So when it comes out in fall 2020 it'll be 3 years.

All comes down to what $500 gets you in fall 2020. I don't see MS decreasing price to $400 or increasing to $600. Then again, you never know.

Xbox X will probably still be around in fall 2020 unless MS cuts it off and goes strictly with Scarlett. X's are still about $400 now, so it'll probably be around $300 by then(?).

So for Scarlett, you basically got:

Scarlett at $400 = ? tf
Scarlett at $500 = ? tf
Scarlett at $600 = ? tf

Just about everyone is ballparking 12 tf for PS5 and Scarlett.
 
Last edited:

FrostyJ93

Member

Sounds great. Panos Paney and his Surface team are likely designi g scarlett for them.
 
Status
Not open for further replies.
Top Bottom