First PS5 Benchmarks Leaked? GTX 1080+ performance?

Take it with a grain of salt and it's likely a devkit but GTX 1080 is where I expected the PS5's GPU to be.

PS5’s Gonzalo APU scored above 20000 points which is very impressive. This is in perspective and compared to the top of the line PC graphics cards.

Source

Pretty solid if true and if it is, likely a dev kit but it's good news nonetheless. Also odd that the 2070 falls below the 1080 by a sizable gap when it's supposed to be slightly better.

How exactly would you benchmark a closed system with no support on other operating systems?

^ This

Couple of red herrings:

1. Fire Strike isn't a benchmark, 3DMark is. Fire Strike is a name of a leaderboard of benchmarks.

2. 3DMark only runs on Windows, Android and iOS. Unless PS5/devkit uses those OS, there's no way to run 3DMark on it.

Sony is really pushing for rumors. Next time maybe don't skip E3, there's no need for this shady stuff.

Isn't around 2070 performance what most have been saying? outside some of the loonies obv

Blue Screen of Death?

@Juub1990:

Fire Strike scores can vary depending on which Fire Stirke edition e.g. Fire Strike, Fire Strike Extreme and Fire Strike Ultra

https://www.3dmark.com/hall-of-fame-2/fire+strike+3dmark+score+performance+preset/version+1.1/1+gpu

@scatteh316: @ten_pints: Didn’t we get PS4’s APU Firestrike benchmark scores?

So the PS5 compares to a GPU that launched in May, 2016... And why not run Firestrike Extreme or Ultra? And how did they run this on a dev kit? What were the PCs CPUs and what were the clocks? So many questions...

Fire Stirke runs on DirectX11.

https://benchmarks.ul.com/3dmark

Fire Strike is a showcase DirectX 11 benchmark for modern gaming PCs. Its ambitious real-time graphics are rendered with detail and complexity far beyond other DirectX 11 benchmarks and games.

@goldenelementxl: Not a lot of details. It’s likely fake but eh.

RX-5700 XT's 225 watts is similar to RX-580's gaming 209 watts power consumption(1), hence both MS and Sony must recycle the lessons learnt from X1X.

https://www.tomshardware.com/reviews/amd-radeon-rx-580-review,5020-6.html

I have no clue what any of this means.

Me neither. I mean 20 000 is a bigger number than 19 370 so I guess that's good lol?

Sometimes it's bad.

PS5 might take 20k seconds to load a game, unlike the others.

Most likely fake since it listed as "over 20,000" LOL. What I am very certain of is the PS5 is going to deliver the performance within reasonable gamers expectations.

I am more interested in the 1080 performance at current system wattage.

@Juub1990 said:

Take it with a grain of salt and it's likely a devkit but GTX 1080 is where I expected the PS5's GPU to be.

PS5’s Gonzalo APU scored above 20000 points which is very impressive. This is in perspective and compared to the top of the line PC graphics cards.

Source

Pretty solid if true and if it is, likely a dev kit but it's good news nonetheless. Also odd that the 2070 falls below the 1080 by a sizable gap when it's supposed to be slightly better.

Yeah i was about to post this.

Keeping my expectations low,but if sony manage to do a console with that kind of power for $500 i would be incredibly surprise.

Is just a spit away from a 2080.

Those still are PC parts dude,i am sure they can test it on a windows setup because even using it on a close system,and in fact the only thing stopping the PS from running windows is sony,the xbox already runs windows this are basically PC part slighty modify.

1-People still call Firestrike the end result.

IN fact do a search for which ever part and put firestrike and the score will pop up.

2-Devkits can use any OS,second this are PC parts i don't get it aren't you suppose to know this? Aren't you a PC gamer? In fact the xbox runs a modify version of windows 10 regardless of been an enclosed platform.

Maybe the part was tested using windows,but is not out of this world that the CHIP get tested,what is been tested here is the PS5 chipo supposedly.

Don't damage control this already,there is no need to is a rumor,this goes for cows as well,wait a little and don't be disappointed later.

In this case the higher number is better. But it's likely fake.

I'm not even a PC Enthusiast to know this as well lol. I'm just a PC gamer and that's about it.

I can answer that since I do own an RTX 2070. GTX 1080 has slightly more cuda cores then the 2070 but for the most parts, 2070 still has an edge over the 1080 depending on the game of course. I went with the 2070 because it's new and I have 3 year warranty if anything goes wrong.

So do you have any proof that sony leaked this?

Around and abode are 2 different things,even if it end equal or a little worse still would be fantastic don't get me wrong.

Now this is some sad damage control the RTX2070 release last year 2018,and this score ""CLAIM"" is slower you know for how much a RTX2070 goes for?

The cheapest model is like $480 the best one like $600+

By Next year the RTX2070 will probably retail for $400 with luck the GPU alone man.

Even if it end been tied or even a little lower than RTX 2070 it should make for a quite a great console.

I expected to enter this thread and see cows dancing for joy,i was ready to tell people to cool their engines,but i find the contrary hermits doing damage control as if this was a fact,puting doubts of how the test was done or even run in the first place,as if the chip inside the PS5 would be like Cell,and not something from PC.

@tormentos: So,a rumor being called out as being suspicious and stating why its suspicious and most likely false, is damage control? We are talking about a rumor that list the propose GPU as "over 20,000" why listing actual numbers for everything else. It really isn't unreasonable to call out the rumor.

This is one hell of a hot take. I didn't know Sony released this report.

Lets ignore that the rumor claim is for the PS5,what would be surprising from that Soc to score 20,000?

Navi 5700XT is say to be faster than Vega 64 which score over 19,000 points.

Now remember that i say ignoring that this was for the PS5,now the so call devkit of the PS5 were say to be over 13TF,but is not 13TF Navi,according to rumors the first devkits had Vega and not the final chip,so a 9+TF Navi could actually outperform a 13TF Vega,which has a firestrike score of 19,200.

So is not that out of this world,but then again that chip even been for the PS5,doesn't mean it will not be water down when it hit the PS5,the GPU in the Pro is much more capable than the 911mhz it has on the PS4 pro.

@tormentos said:

@phbz said:

Sony is really pushing for rumors. Next time maybe don't skip E3, there's no need for this shady stuff.

So do you have any proof that sony leaked this?

@howmakewood said:

Isn't around 2070 performance what most have been saying? outside some of the loonies obv

Around and abode are 2 different things,even if it end equal or a little worse still would be fantastic don't get me wrong.

@goldenelementxl said:

So the PS5 compares to a GPU that launched in May, 2016... And why not run Firestrike Extreme or Ultra? And how did they run this on a dev kit? What were the PCs CPUs and what were the clocks? So many questions...

Now this is some sad damage control the RTX2070 release last year 2018,and this score ""CLAIM"" is slower you know for how much a RTX2070 goes for?

The cheapest model is like $480 the best one like $600+

By Next year the RTX2070 will probably retail for $400 with luck the GPU alone man.

Even if it end been tied or even a little lower than RTX 2070 it should make for a quite a great console.

I expected to enter this thread and see cows dancing for joy,i was ready to tell people to cool their engines,but i find the contrary hermits doing damage control as if this was a fact,puting doubts of how the test was done or even run in the first place,as if the chip inside the PS5 would be like Cell,and not something from PC.

Next year Nvidia is also bringing 7nm Samsung fab Ampere cards out so nobody is going to even look at 400$ 2070

Considering the Pricing that has been going on for some years i don't expect the RTX2070 to fall well below $400 at anytime soon.

The 1070 was basically held at that price with some deals having it for time to time in $389 or something like that but for the most part it was over $400.

You will not see nvidia dropping prices of their GPU to rotten meat at all,that is now how they operate.

They have the lead by a considerable margin,own the most powerful GPU so they hold the card.

If something like this is true for the PS5,don't expect DF to build a potato machine for less than the PS5 or close to beat it.

As I mentioned earlier, what I am more interested in for both consoles is the power consumption. The Xbox One X and PS4 Pro consumes between 100-105W. The upcoming Navi 5700 XT blows throw this. So, how are both companies going to meet the current power constraints?

Sup there, 2070 buddy. I never had experience with the 1080 myself, but I went straight to an RTX card when the opportunity presented itself.

Only thing is, I'm having weird refresh rate issues on couple of games at 1440p that I don't have at 4K or 1080p.

Well i am just telling you that the score been 20,000 is not out of this world,for been Navi the GPU in question,but i my self know that even if that is the real silicon that doesn't mean it will be exactly as powerful,it can be weaker like the Pro was vs other GPU with 36CU.

When the 5700 XT is the hands of reviewers then I can comment on the obviously fake score regardless of how close the actual performance maybe.

For those who wonder how can they test it.

The PS4 scored 5000 the 7850 has like 4700 score.

@tormentos said:

For those who wonder how can they test it.

The PS4 scored 5000 the 7850 has like 4700 score.

That's a 4x improvement over the PS4 in terms of GPU. That seems pretty low to me but I don't know the specifics of the tests.

My sources confirmed me that the real score was 29999 but Sony was super modest.

From now on, and to keep things clear, we should just measure a platforms performance in kuriks. That way everyone is on the same playing field, and we can instantly tell who wins.

@Zero_epyon said:

@tormentos said:

For those who wonder how can they test it.

The PS4 scored 5000 the 7850 has like 4700 score.

That's a 4x improvement over the PS4 in terms of GPU. That seems pretty low to me but I don't know the specifics of the tests.

low lol really?

Let's just put this in perspective for anyone thinking that:

• RTX 2070 scored 18,103 ; MSRP $499
• Vega 64 scored 19, 200 ; MSRP $499
• RTX 2080TI scored 21, 892 ; MSRP $1,199

So that puts PS5's gonzalo score of 20,000+ in between the performance rating of RTX 2070 and RTX 2080TI with something approximate to what a RTX 2080 would be; that's a $799 MSRP GPU alone.

These are currently the top level GPUs money can buy even for consumer PC hardware.

Does it still "Seems pretty low" to you? ¯\_(ツ)_/¯

Unless you were expecting a 2080TI in a $500 console, those projections far exceed what most people could have hoped for.

@Zero_epyon said:

@tormentos said:

For those who wonder how can they test it.

The PS4 scored 5000 the 7850 has like 4700 score.

That's a 4x improvement over the PS4 in terms of GPU. That seems pretty low to me but I don't know the specifics of the tests.

That was my thinking as well,but then you see the RX2070 scored in the low end of 18,000 which yet is not even 4 times higher than the PS4,yet the performance delta is huge you see that is ok.

Some one is bitter.

Man is a rumor don't damage control it so hard.

@boxrekt: Didn't you say that next gen is going to be the biggest generational jump? If that's the case 4x is closer to midgen upgrade than next gen upgrade. :)

@tormentos said:

@Zero_epyon said:

@tormentos said:

Some one is bitter.

Man is a rumor don't damage control it so hard.

Sounds more like a leak than a rumor to me.

Apisak, didn't say, "there is a rumored benchmark.". He specifically said, QUOTE: "There is also a new benchmark for Gonzalo". no disclaimer.

So he is looking at the real numbers for that APU. The only question would be if that's the actually Gonzalo's APU he's looking at and if so is PS5 using Gonzalo.

GTX 1080 will be 4.5 years old by Holiday 2020. Not impressed at all if this is around the performance level the consoles will deliver.

Taking it kinda it hard huh Pedro? Didn't you say consoles would have not graphical leap only a slight upgrade?

Now you hear it will be near RTX 2080 performance. How will you be able to sleep at night?

Dry your baby man tears

MS will still provide you with plenty of mediocre shovelware via continuous paid streaming apps just like you like it.

Everyone wins :)

@boxrekt said:

@Pedro said:

@boxrekt: Didn't you say that next gen is going to be the biggest generational jump? If that's the case 4x is closer to midgen upgrade than next gen upgrade. :)

Taking it kinda it hard huh Pedro? Didn't you say consoles would have not graphical leap only a slight upgrade?

Now you hear it will be near RTX 2080 performance. How will you be able to sleep at night?

Dry your baby man tears

MS will still provide you with plenty of mediocre shovelware via continuous paid streaming apps just like you like it.

Everyone wins :)

Pedro still fails to understand that X1X and PS4 Pro games are designed with base ps4/x1 hardware in mind and that they do not take all advantage of the midgen upgrades... Show him a benchmark for the base consoles and MAYBE he'll understand, but don't expect much, this is the same guy that used to say X1X and PS4 Pro would run NextBox and PS5 games on lower settings and that generations were over...

@boxrekt said:

@Zero_epyon said:

@tormentos said:

For those who wonder how can they test it.

The PS4 scored 5000 the 7850 has like 4700 score.

That's a 4x improvement over the PS4 in terms of GPU. That seems pretty low to me but I don't know the specifics of the tests.

low lol really?

Let's just put this in perspective for anyone thinking that:

• RTX 2070 scored 18,103 ; MSRP $499
• Vega 64 scored 19, 200 ; MSRP $499
• RTX 2080TI scored 21, 892 ; MSRP $1,199

So that puts PS5's gonzalo score of 20,000+ in between the performance rating of RTX 2070 and RTX 2080TI with something approximate to what a RTX 2080 would be; that's a $799 MSRP GPU alone.

These are currently the top level GPUs money can buy even for consumer PC hardware.

Does it still "Seems pretty low" to you? ¯\_(ツ)_/¯

Unless you were expecting a 2080TI in a $500 console, those projections far exceed what most people could have hoped for.

Stop spreading lies. The RTX 2080 Ti is not even listed in the benchmark. The regular RTX 2080 scored 21,892 and its launch price was $799 USD.

Devkits can't use any OS they want, especially if you have to license the said OS. OS can't run magically on anything, you need to have a certified BIOS and drivers. Ever wondered why macOS don't run on the same 'PC' hardware without hacks or why Mac's didn't run Windows before Apple adding support for dual boot? Of course idiot like you who doesn't know anything about anything apart from google things.

Now go shooo....try to google some random unrelated article for a comeback....shoooo.

Is a spit from the 2080RTX, Gonzalo has 20,000 vs the RTX2080 21,872 which released last year.

In fact Gonzalo - 20,000 vs RTX 2070 - 18,103 difference 1,897

RTX 2080 - 21,892 vs Gonzalo - 20,000 difference 1,892

Now that shit was funny! Just read a thread with Tormy and Ron, now that is a real damn treat of complete shit.

@BassMan said:

@boxrekt said:

@Zero_epyon said:

@tormentos said:

For those who wonder how can they test it.

The PS4 scored 5000 the 7850 has like 4700 score.

That's a 4x improvement over the PS4 in terms of GPU. That seems pretty low to me but I don't know the specifics of the tests.

low lol really?

Let's just put this in perspective for anyone thinking that:

• RTX 2070 scored 18,103 ; MSRP $499
• Vega 64 scored 19, 200 ; MSRP $499
• RTX 2080TI scored 21, 892 ; MSRP $1,199

So that puts PS5's gonzalo score of 20,000+ in between the performance rating of RTX 2070 and RTX 2080TI with something approximate to what a RTX 2080 would be; that's a $799 MSRP GPU alone.

These are currently the top level GPUs money can buy even for consumer PC hardware.

Does it still "Seems pretty low" to you? ¯\_(ツ)_/¯

Unless you were expecting a 2080TI in a $500 console, those projections far exceed what most people could have hoped for.

Stop spreading lies. The RTX 2080 Ti is not even listed in the benchmark. The regular RTX 2080 scored 21,892 and its launch price was $799 USD.

ok calm down, I misread that. No need to get pissy

Still doesn't change my point.

if the APU is in between RTX2070 and RTX 2080 that's still incredible for a stand alone console. RTX 2070 and 2080 are $499 and $799 respectively.

And yet Apisak called it a benchmark!

Who do we believe a trusted insider PC parts leaker who has accurately leaked similar information in the past not to mention the actual Gonzalo SOC or professional do nothing system wars forum fanboy pc_rocks?