Support NeoGAF

[Gamernyc78]

To act as if Xbox didn't catch up to Sony in the 360 PS3 generation is just being disingenuous. That is also a mighty strong reaction to someone suggesting a path for MS to be competitive again.

Catch up? Thy gained ground yes as a new contender gen to next Gen but there was no catching up to Sony as Xbox was released a year and change ahead of ps3 and ps3 still caught up. Any ground thy gained on their market share thy dwindled away completely with Xbox one. Microsoft never did nor will thy catch up to Sony. Now them gaining back the ground thy gained and lost this gen sure it might be possible especially in the U. S. nothing disgenuous about what I'm saying.

Anyway bro this topic has been talked to death through all these years and it's boring. You have your view and I have mine but I'm leaving it alone.

 

[TLZ]

Navi 2X not being Navi 10 Lite, which Oberon is, is the last nail in the GitHub leaks coffin.

What specs do we know of the NAvi 2x series btw? I remember seeing one ages ago but can't remember much.

 

[Gediminas]

To act as if Xbox didn't catch up to Sony in the 360 PS3 generation is just being disingenuous. That is also a mighty strong reaction to someone suggesting a path for MS to be competitive again.

disingenuous would be count those broken xbox consoles, which where on every corner, which blew out sale numbers by millions + their console was hacked, so they sold more because of that. i know, i have one. games costed like 3 euro after that. and in my country almost all gamers had an xbox, but non of them legit games. and still, PS3 outsold them. + we can look at xbox one sales, which are 41-44M. so, there was no catch in up.

 

[Matreios Raven]

JFC, so now it’s easy for MS to reach 14 TF but PS can’t reach 12. Talk about changing the goalposts.

 

[TLZ]

I agree wholeheartedly that's your opinion but many have their own as well. Mine being that it's below average, the amount of R&D Microsoft spent and no to mention marking. The effort does not make the results.

Sure of course. But big failure to me is for something under 20 mil.

 

[Gamernyc78]

What i am suggesting is. Microsoft felt they needed to work with AMD to co-author and develop a new API to obtain the best performance from the hardware ray tracing.

Sony doing similar work on its own, it very unlikely they’ll receive the same gains? This just speculation as we don’t know what Sony API development is for the hardware ray tracing. It’s interesting AMD and Microsoft co designed an API specifically for ray tracing for their console.

And I don't agree in the slightest tht Sony doing the API on its own won't have the same or better gains. This was the same talk during the dx12 with again many insiders and ppl talking about xbox advantage in that field and Sony created their own apis which were as good or better.

Sony has some geniuses over there tht know their shit.

 

[SmokSmog]

lol 9 rdna 1.0 tflops is 6 rdna 2.0 tflops.

think about it. github truthers were saying the ps5 was going to be 6 tflops lmao.

RDNA2 doesn't have 50% ipc uplift but 50% perf per wat this is not the same
Example

12TF RDNA1 150w 100fps/s
12TF RDNA2 100W 100fps/s
RDNA2 50% better perf/W than RDNA1



12TF RDNA1 150w 100fps/s
12TF RDNA2 150W 150fps/s

RDNA2 50% better IPC than RDNA 1

 

[Zero707]

Did you not understand what I just said, or...

AMD confirmed both are using the latest Zen2 architecture, and both are using the newest RDNA2 Navi2x.

The GitHub was claimed to be Navi10 by ro_game himself. That rules that part of the data out.

the amount of cores don't tell us about the architecture, all we know today thats Navi2x is the codename for RDNA 2 and not for big navi, you will see laptop based on rdan 2, all i'm saying Consoles are custom and are different than offshelf market

 

[DeepEnigma]

the amount of cores don't tell us about the architecture, all we know today thats Navi2x is the codename for RDNA 2

So all the worshipping people did to _rogame for breaking it all down., is now goalposted?

FFS, the narrative was Navi10 until today with the rest of the data.

 

[icerock]

how does this make github dead?

to quote Colbert

The Github leak had data about BC testing (PS5). Stating a full count of CUs. A 2000 Mhz mode is documented in there as well which would make more sense for a smaller chip.

There was no info about an architecture. The architecture is a missing information and I actually don't think it would have any relevance to be in there, just like the info about RT and VRS because BC was tested and not something else.

That chip being RDNA 2 makes actually more sense as those 2000 Mhz are easier to achieve by it.

Edit:
That silicon is semi custom anyway, so it is actually something different than pure RDNA 2 would be.

Once again, this is not true. Pure conjecture, the Github data stated full count of CUs for that particular testing. Which he admits, in his own words are related to BC compatibility. Hence using 9WGP to mimic PS4, and 18WGP using to mimic PS4 Pro. Nobody can tell how many CUs were disabled, it could be 4, it could be more. This is more mental gymnastics to try to box into a theory one so desperately wants to believe, despite not many pieces fitting into the puzzle.

Also, if that chip is indeed RDNA 2, it doesn't need to be chained to the design/limits of 5700 series products right? So once again, assuming that the chip possesses all the traits of RDNA 2, and had VRS along with RT, can anyone shine a light on the overall core count on the Oberon A0 iGPU in that repo?

Answer, is nobody can, because the data doesn't contain any information on it. Anybody who's claiming they do are just shooting in the dark.

 

[thismeinteil]

To act as if Xbox didn't catch up to Sony in the 360 PS3 generation is just being disingenuous. That is also a mighty strong reaction to someone suggesting a path for MS to be competitive again.

MS was only competitive when Sony shot themselves in the foot with a $200 price difference and complicated architecture that lead to worse multiplats for the first year or two. Even then, Sony beat them and came less than 15M away from the leader that gen, the Wii. Sony isn't making those mistakes, again.

 

[MarkMe2525]

Catch up? Thy gained ground yes as a new contender gen to next Gen but there was no catching up to Sony as Xbox was released a year and change ahead of ps3 and ps3 still caught up. Any ground thy gained on their market share thy dwindled away completely with Xbox one. Microsoft never did nor will thy catch up to Sony. Now them gaining back the ground thy gained and lost this gen sure it might be possible especially in the U. S. nothing disgenuous about what I'm saying.

Anyway bro this topic has been talked to death through all these tears and it's boring. You have your view and I have mine but I'm leaving it alone.

Yes 360 came out a year early. Are you not aware that PS3 sold for a year longer. You wrote a paragraph framing the idea that Xbox did not catch up to PS3 during that generation and like I stated, that is disingenuous
Btw I could absolutely care less who sells more. It really makes no sense why someone would.

 

[thicc_girls_are_teh_best]

Killed by their own god

Navi 2X not being Navi 10 Lite, which Oberon is, is the last nail in the GitHub leaks coffin.

You guys should really, REALLY read the post I made a page back. Particularly the quoted part. Explains very well what the Oberon tests actually are for, i.e Ariel iGPU regression profile testing. The tests have not actually tested the Oberon chip outside of the Ariel iGPU profile from what we can see.

That would explain the Navi 10 listing in Rogame's post: it's referencing the Ariel iGPU profile. Same thing with the "full chip" listing in the first Oberon leak (whichever one said "full chip", anyway). This is actually a best-case scenario for PS5 because it means we haven't gotten a full look on the Oberon chip, even the E0 version which is most recent.

Let's hope a benchmark or testing data for Oberon outside of Ariel iGPU profile surfaces soon, it would answer many currently unanswered questions (and also show RT/VRS/VRR etc. enabled I'd assume).

Once again, this is not true. Pure conjecture, the Github data stated full count of CUs for that particular testing. Which he admits, in his own words are related to BC compatibility. Hence using 9WGP to mimic PS4, and 18WGP using to mimic PS4 Pro. Nobody can tell how many CUs were disabled, it could be 4, it could be more. This is more mental gymnastics to try to box into a theory one so desperately wants to believe, despite not many pieces fitting into the puzzle.

Also, if that chip is indeed RDNA 2, it doesn't need to be chained to the design/limits of 5700 series products right? So once again, assuming that the chip possesses all the traits of RDNA 2, and had VRS along with RT, can anyone shine a light on the overall core count on the Oberon A0 iGPU in that repo?

Answer, is nobody can, because the data doesn't contain any information on it. Anybody who's claiming they do are just shooting in the dark.

Virtually this. Plus, I just think a lot of people are misinformed what relation Ariel actually has to Oberon (they are not the same chip) and what the Oberon testing data was actually configured to (Ariel iGPU profile, for regression).

Assuming strong chance there are more disabled CUs on Oberon (up to perhaps as many CUs for the chip in total as Arden; E0 did not provide a lot of other info besides increased bandwidth which, tbf, still indicates a 256-bit bus), the only thing really in danger of going obsolete now is PS5 being 9.2TF; if Oberon's a bigger chip in reality it is then most certainly more than 9.2TF.

 

[DeepEnigma]

JFC, so now it’s easy for MS to reach 14 TF but PS can’t reach 12. Talk about changing the goalposts.

We didn’t have 50% improvement to efficacy over RDNA until today. So people are entertaining that. What’s your problem?

 

[Old Empire.]

And I don't agree in the slightest tht Sony doing the API on its own won't have the same or better gains. This was the same talk during the dx12 with again many insiders and ppl talking about xbox advantage in that field and Sony created their own apis which were as good or better.

Sony has some geniuses over there tht know their shit.

DX12 helped indeed, it is the reason lot of xbox games in later cycle maintained the frame rate. Xbox resolution also jumped to 900p + and 1080p with some games.

Xbox games with D11 was mostly 720p.

PS4 has stronger performing memory and better GPU. DX12 can’t close the gap everywhere. It had nothing to do with Sony devs performing miracles, PS4 was better console spec wise.

 

[Zero707]

So all the worshipping people did to _rogame for breaking it all down., is now goalposted?

FFS, the narrative was Navi10 until today with the rest of the data.

ehh i didn't comment on github untill recently so yeah i hope Oberon will have more than 36CU

 

[DeepEnigma]

You guys should really, REALLY read the post I made a page back. Particularly the quoted part. Explains very well what the Oberon tests actually are for, i.e Ariel iGPU regression profile testing. The tests have not actually tested the Oberon chip outside of the Ariel iGPU profile from what we can see.

That would explain the Navi 10 listing in Rogame's post: it's referencing the Ariel iGPU profile. Same thing with the "full chip" listing in the first Oberon leak (whichever one said "full chip", anyway). This is actually a best-case scenario for PS5 because it means we haven't gotten a full look on the Oberon chip, even the E0 version which is most recent.

Let's hope a benchmark or testing data for Oberon outside of Ariel iGPU profile surfaces soon, it would answer many currently unanswered questions (and also show RT/VRS/VRR etc. enabled I'd assume).

I agree with you here, that’s what a lot of people have been saying since the leak. That we are not seeing the whole picture. From Matt to Klee and everyone using logic in between.

 

[Gamernyc78]

Yes 360 came out a year early. Are you not aware that PS3 sold for a year longer. You wrote a paragraph framing the idea that Xbox did not catch up to PS3 during that generation and like I stated, that is disingenuous
Btw I could absolutely care less who sells more. It really makes no sense why someone would.

Smh are you aware tht ps3 sold more in the end; hundreds of thousands if not millions of rrod affected units caused ppl to buy new ones anddddd my favoriteeee, launch numbers aligned ps3 at hundreds more sold more almost every month world wide?

Bro this isn't fanboy shit but facts. It's not about caring but Xbox never caught up to Sony and tht is my point.

 

[MarkMe2525]

MS was only competitive when Sony shot themselves in the foot with a $200 price difference and complicated architecture that lead to worse multiplats for the first year or two. Even then, Sony beat them and came less than 15M away from the leader that gen, the Wii. Sony isn't making those mistakes, again.

Yes... And MS caught up because of that. I agree that Sony probably won't make those mistakes again. I don't believe anyone was saying they were. That is really awesome that Sony almost caught the Wii. Why does that matter? Thank you for your statement.

 

[SlimySnake]

RDNA2 doesn't have 50% ipc uplift but 50% perf per wat this is not the same
Example

12TF RDNA1 150w 100fps/s
12TF RDNA2 112W 100fps/s
RDNA2 50% better perf/W than RDNA1



12TF RDNA1 150w 100fps/s
12TF RDNA2 150W 150fps/s

RDNA2 50% better IPC than RDNA 1

ok fine.

but that means sony is going with a 9.2 tflops gpu for 106w instead of going with a 12 tflops gpu for 97w because they didnt have the good sense to go wide and slow like MS.

 

[MarkMe2525]

Smh are you aware tht ps3 sold more in the end; hundreds of thousands if not millions of rrod affected units caused ppl to buy new ones anddddd my favoriteeee, launch numbers aligned ps3 at hundreds more sold more almost every month world wide?

Bro this isn't fanboy shit but facts. It's not about caring but Xbox never caught up to Sony and tht is my point.

Yes I am aware that Sony sold more. I don't think anyone said differently. Yes that isn't fanboy shit but facts. That also doesn't change the "fact" that MS did really well that gen and caught up with Sony in sales. Why does it seem like you are upset about this?

 

[StreetsofBeige]

disingenuous would be count those broken xbox consoles, which where on every corner, which blew out sale numbers by millions + their console was hacked, so they sold more because of that. i know, i have one. games costed like 3 euro after that. and in my country almost all gamers had an xbox, but non of them legit games. and still, PS3 outsold them. + we can look at xbox one sales, which are 41-44M. so, there was no catch in up.

Disingenuous would be to count those broken YLOD PS3 consoles, which were on every corner, which blew out sales numbers by millions + PS3 also acted as a cheap BR player, so lots of HD movie watchers and people who wanted to kill two birds with one stone got a PS3 for movies too.

That's why even though both systems had roughly the same number of total systems sold, third party games sold better on 360 than PS3.

There is no catching up of Xbox One replicating 360, just as there's no catching up from PS3 or PS4 against PS2 from 15 years ago which sold about 150 million units, while PS3 and PS4 have sold about 50-60 million fewer units each.

The gaming industry is exponentially bigger than 25 years ago, yet PS4 has sold about the same number of units as PS1 despite the PS4 generation going on longer too.

 

[SmokSmog]

ok fine.

but that means sony is going with a 9.2 tflops gpu for 106w instead of going with a 12 tflops gpu for 97w because they didnt have the good sense to go wide and slow like MS.

yea they should get 12-14TF with 50% better pef/w with wide gpu.

 

[Gamernyc78]

DX12 helped indeed, it is the reason lot of xbox games in later cycle maintained the frame rate. Xbox resolution also jumped to 900p + and 1080p with some games.

Xbox games with D11 was mostly 720p.

PS4 has stronger performing memory and better GPU. DX12 can’t close the gap everywhere. It had nothing to do with Sony devs performing miracles, PS4 was better console spec wise.

No one said dx12 didnt help, both Sony and Microsoft apis helped but it was no advantage. The point is Sony ppl are competent to make an API or solution tht is on par or better. Ps4 being better spec wise did make games better but your assuming tht had both consoles been equal Sony's ganes wouldn't have performed the same as xbox lol no evidence whatsoever to tht point and dx12 was doing what Sony's API was doing nothing better. Alluding to tht us disgenuous. Dx12 was suppose to be this game changer it game changed shit.

With time and as years go by most new games push or perform better on all consoles because devs have been learning how to deal with the hardware and obviously using apis. It's not exclusive to Microsoft lol

 

[MarkMe2525]

disingenuous would be count those broken xbox consoles, which where on every corner, which blew out sale numbers by millions + their console was hacked, so they sold more because of that. i know, i have one. games costed like 3 euro after that. and in my country almost all gamers had an xbox, but non of them legit games. and still, PS3 outsold them. + we can look at xbox one sales, which are 41-44M. so, there was no catch in up.

Ok........ Those are all valid reasons. Not the ones I would use to make a point, but valid. Actually I'm not sure what point you are trying to make.

 

[grfunkulus]

Lol, same company that showed off a bunch of bogus charts with the 3600 ahead of the 9900k in gaming performance.

Anyone that thinks you'll have a retail product on shelves in 6 months that is 50% better than a 2080ti at $700 is obviously a pie in the sky dreamer.

 

[Mod of War]

Off Topic so forgive me and show me mercy! But have you heard anything about a Stargate SG1 reboot?

I heard they’re working on reviving and Richard Dean Anderson and Amanda Tapping might be involved. We should make a thread in OT for rumors and the like in it.

As for the rest of the thread, take the PS360 sales talk out of here and move it to a new thread for the 100th time on that tiring subject.

 

[B_Boss]

Jesus tapdancin' Christ, that Colbert dude is insufferable.

How many times in how many ways can Github literally be shot down as not entirely accurate of anything, but you refuse to give up?

This all reminds me of The Last Samurai, wherein the Github clingers are being smacked down by Ujio with a wooden sword at every turn, but refuse to stay down.

Until Ujio gave his ass the final smack down....and Algren stayed down .

 

[James Sawyer Ford]

Jesus tapdancin' Christ, that Colbert dude is insufferable.

How many times in how many ways can Github literally be shot down as not entirely accurate of anything, but you refuse to give up?

This all reminds me of The Last Samurai, wherein the Github clingers are being smacked down by Ujio with a wooden sword at every turn, but refuse to stay down.

Because GitHub agrees with the 321st revision of his chart

 

[MARTYWOLF]

PS5 is really 4tf Rdna2 like lockhart. Both balanced systems

Just joking bro but it's confirmed ps5 is Rdna2 like nextbox.

Hahaha sweet as guess nothing on Lockhart or the potential ps5 pro/base. Rdna2 will definitely help with power consumption! Only got 4 hrs left of work than I can read all the Radeon goods

 

[MadAnon]

So if rDNA 2 is 50% better ppw then RDNA1 does that mean XSX is the equivalent of a 24tf gcn?

Wow if so.

It's performance per watt. Something completely different. It means same performance but at lower power draw compared to RDNA.

 

[thismeinteil]

Yes... And MS caught up because of that. I agree that Sony probably won't make those mistakes again. I don't believe anyone was saying they were. That is really awesome that Sony almost caught the Wii. Why does that matter? Thank you for your statement.

It just shows how strong the PS brand is. MS had disadvantages with worse multiplats this gen, but they only had a $100 difference, which also included a camera. But, where it took Sony 3 years to get the price down to acceptable levels for the PS3, MS did it in just a few months after launch this gen. PS4 will still go on to sell 120M+, while XBO will finish around 50M-55M. Xbox just isn't as strong of a brand to carry it through missteps. Definitely not if Sony is firing on all cylinders.

 

[DeepEnigma]

They will be on the 7nm+ as well.

AMD Financial Day 2020 [OT] PS. 5 Goddamn Hours Long

Thread continues to deliver!

www.neogaf.com

 

[Matreios Raven]

We didn’t have 50% improvement to efficacy over RDNA until today. So people are entertaining that. What’s your problem?

My issue is by default, only MS can reach 12-14 TF, but Sony can’t?

 

[icerock]

You guys should really, REALLY read the post I made a page back. Particularly the quoted part. Explains very well what the Oberon tests actually are for, i.e Ariel iGPU regression profile testing. The tests have not actually tested the Oberon chip outside of the Ariel iGPU profile from what we can see.

That would explain the Navi 10 listing in Rogame's post: it's referencing the Ariel iGPU profile. Same thing with the "full chip" listing in the first Oberon leak (whichever one said "full chip", anyway). This is actually a best-case scenario for PS5 because it means we haven't gotten a full look on the Oberon chip, even the E0 version which is most recent.

Let's hope a benchmark or testing data for Oberon outside of Ariel iGPU profile surfaces soon, it would answer many currently unanswered questions (and also show RT/VRS/VRR etc. enabled I'd assume).



Virtually this. Plus, I just think a lot of people are misinformed what relation Ariel actually has to Oberon (they are not the same chip) and what the Oberon testing data was actually configured to (Ariel iGPU profile, for regression).

Assuming strong chance there are more disabled CUs on Oberon (up to perhaps as many CUs for the chip in total as Arden; E0 did not provide a lot of other info besides increased bandwidth which, tbf, still indicates a 256-bit bus), the only thing really in danger of going obsolete now is PS5 being 9.2TF; if Oberon's a bigger chip in reality it is then most certainly more than 9.2TF.

Fantastic post, thanks for sharing, it makes a lot of sense. Oberon is the PS5 iGPU, we just don't know the full make up of the damn thing.

But, something which has also bothered me for a while is when was Oberon A0 manufactured? As in, what is the time lapse between the date we saw earliest findings of existence of Oberon A0? Because Oberon E0 wasn't discovered until January 2020, which to me means it's a mid-late 2019 product.

BTW, I still don't know what to make of power mantra. If MS are stopping themselves at 1.7GHz, it must be for a reason. I find it incredibly hard to believe, that engineers at Sony magically found a way to extract 300MHz more clock speed out of these iGPUs on an architecture which is more efficient, unless they decided fuck it, let's spend the moolah on HBM2 memory to control the power consumption and push these clocks to the max. I still feel the Github was a mixture of BC compatibility/regression/stress testing and those 'native' clock speeds aren't indicative of the final product. 2.0GHz clock in a console is absolutely baloney to me, I reckon the final rendition of Oberon has a bigger CU count but a more modest clock speed, [email protected] gets them just over 12TF mark and basically pegs the two level.

 

[MadAnon]

Once again, this is not true. Pure conjecture, the Github data stated full count of CUs for that particular testing. Which he admits, in his own words are related to BC compatibility. Hence using 9WGP to mimic PS4, and 18WGP using to mimic PS4 Pro. Nobody can tell how many CUs were disabled, it could be 4, it could be more. This is more mental gymnastics to try to box into a theory one so desperately wants to believe, despite not many pieces fitting into the puzzle.

Also, if that chip is indeed RDNA 2, it doesn't need to be chained to the design/limits of 5700 series products right? So once again, assuming that the chip possesses all the traits of RDNA 2, and had VRS along with RT, can anyone shine a light on the overall core count on the Oberon A0 iGPU in that repo?

Answer, is nobody can, because the data doesn't contain any information on it. Anybody who's claiming they do are just shooting in the dark.

36CUs at 2ghz were specifically refered to as "full chip" performance. This "we don't know how many CUs were disabled" narrative needs to die already.

Obviously doesn't mean It's 100% PS5 chip.

 

[Shambala]

Yes 360 came out a year early. Are you not aware that PS3 sold for a year longer. You wrote a paragraph framing the idea that Xbox did not catch up to PS3 during that generation and like I stated, that is disingenuous
Btw I could absolutely care less who sells more. It really makes no sense why someone would.

Bro it didn’t sell for a year longer. If it came out a year earlier therefore they sold for approximately the same time

 

[DeepEnigma]

My issue is by default, only MS can reach 12-14 TF, but Sony can’t?

Who’s saying that?

 

[FacelessSamurai]

I agree wholeheartedly that's your opinion but many have their own as well. Mine being that it's below average, the amount of R&D Microsoft spent and no to mention marking. The effort does not make the results.

It’s been confirmed to be a rounded number, Xbox one above 50 million. Lower than 360 for sure though.

 

[StreetsofBeige]

All this talk about systems running at 1.7 - 2 ghz

Way back, remember when Intel had 4 ghz cpus?!

 

[MarkMe2525]

Bro it didn’t sell for a year longer. If it came out a year earlier therefore they sold for approximately the same time

Not accurate, mod of War asked that this conversation not continue in thread. I PM'd you their production run dates.

 

[Moridianae]

Yeah, I'm more inclined to say that the 12-12.9 TF being touted is a reasonable expectation with the efficiency capabilities of this tech.

Still---12 to 13TF is perfectly achievable here. It is looking more likely that the systems will be very, very similar.

I think Sony hits 12TF for even ground or 13 to say "the most powerful console"(which doesn't mean anything here)

The reason I say they go equal or slightly better is because 12TF has been screamed from the rooftops by MS, so any less than that would be seen as "bad" image wise, even though we know its negligible.

 

[thismeinteil]

I knew better than to engadge.... Yet I still did
Dude... What on Earth are you talking about?

You asked a question. I answered it. Simple as that.

 

[icerock]

36CUs were specifically refered to as "full chip" performance. This "we don't know how many CUs were disabled" narrative needs to die already.

Obviously doesn't mean It's 100% PS5 chip.

Did you even read my post in its entirety? I know full well 18WGP means 36CUs and they were tagged with "full chip" performance. But, need I remind you the purpose of that testing? Oh wait, it was for BC compatibility. So, tell me again, why would they activate more than 9WGP and 18WGP when the purpose was to test how the iGPU is mimicking PS4 and PS4 Pro?

Also, "we don't know many CUs were disabled" isn't a narrative, its pure common sense. There isn't going to be an iGPU with 36/36 active CUs, because guess what, yields matter, and these companies don't have bottomless pit of money to manufacture their APUs.

I suggest you to inform yourself more on these things before delving into a discourse, not least offer wisdom to those who are trying to offering a perspective of what this data actually entails.

 

[thicc_girls_are_teh_best]

Last thing I got for tonight. Something interesting came to me: if both systems are targeting 12TF (we at least know one has hit that target but let's just keep it simple say that's what they'll both be), and RDNA2 is 50% more efficient over RDNA1, and the 2080TI is 34% stronger than 5700XT (9.7TF) which is an RDNA1 GPU...

In that case next-gen could be 16% better than 2080TI. And the 2080TI is already 13.45TF, so 12TF RDNA2 should come in around 13.8TF - 13.92TF relative to 2080TI!!

People might've been aiming too low thinking "only" regular 2080 would get lapped here. Very impressive RDNA2 is looking to be

And again, RDNA2 kinda needs it 'cuz like the video says, it's targeting Turing, not Ampere. That's likely what RDNA3 will be for.

 

[MarkMe2525]

Not accurate, mod of War asked that this conversation not continue in thread. I PM'd you their production run dates.

My statement is wrong. I misread his message

 

[MadAnon]

Did you even read my post in its entirety? I know full well 18WGP means 36CUs and they were tagged with "full chip" performance. But, need I remind you the purpose of that testing? Oh wait, it was for BC compatibility. So, tell me again, why would they activate more than 9WGP and 18WGP when the purpose was to test how the iGPU is mimicking PS4 and PS4 Pro?

Also, "we don't know many CUs were disabled" isn't a narrative, its pure common sense. There isn't going to be an iGPU with 36/36 active CUs, because guess what, yields matter, and these companies don't have bottomless pit of money to manufacture their APUs.

I suggest you to inform yourself more on these things before delving into a discourse, not least offer wisdom to those who are trying to offering a perspective of what this data actually entails.

Dude, seriously? They also tested 3rd native mode with 36CUs at 2ghz which was refered to as "full chip" performance. If it was just BC tests then explain that native "full chip" mode. Maybe go read those files again. They were not just BC tests.

There won't be 36/36CU GPU? No shit, sherlock.

 

[TherapistGaymer]

we should be more careful with these kind of messages. it directly implies corruption, dishonesty and bias from DF members.

Sorry man this is neogaf not highschool or the work place. People say whatever.

 

[DeepEnigma]

All this talk about systems running at 1.7 - 2 ghz

Way back, remember when Intel had 4 ghz cpus?!

These are GPUs not CPUs, and AMD does have CPUs that run over 4Ghz.

Console CPUs will probably be 3.2-3.5Ghz.

 

[Zero707]