By using this site, you agree to our Privacy Policy and our Terms of Use. Close

Forums - Nintendo Discussion - Maxwell (NX) vs GCN (PS4/XBOne): Real World Teraflop Performance Comparison.

Soundwave said:
bonzobanana said:
People seem to be forgetting the ps4 and xbox one are supported by large capacity optical drives and have internal hard drives for good storage. The NX is a tablet with games coming on lower capacity cartridges and no hard drive. The ambition of NX games will be closer to portable games than home consoles for that reason. NX games will not compare to ps4 and xbone for other reasons than just gpu or cpu performance. If Nintendo really do use very large capacity cartridges we will pay dearly for those games and they will be uncompetitive with ps4 and xbone pricing.

The comparison of NX versus PS4 or xbone will be like wii vs 360 and ps3. What few multiformat games the NX gets will be cut down versions.

It might be an impressive and interesting tablet system but will be a poor home console.

Anyway do we really want to go backwards again with Nintendo supporting a weak system that performs below that of established console products with large software libraries and are just being replaced with much stronger home consoles anyway which are massively superior to NX.

It's highly likely the NX will be dead on arrival unless it really has a gimmick that can save it. That gimmick is going to be far more important than Tegra X1 in NX's success.

Try putting your PS4 in your back pack and playing it on the go. 

Or playing Android apps on your PS4. 

There's your "gimmick". 

32GB (and eventually 64GB) is the same size as a Blu-Ray disc too, so I don't think that's that big of a deal. Even the Nvidia Shield microconsole could accept a 2.5 inch HDD inside its casing too. 

are you just making it all up?

1- What does the NX have to do with putting the ps4 on the back pack and playing on the go? PS4 won't run the same level of games as the NX anymore than they run the same type of games as the Vita. Sure some smaller games probably would work but no AAA games, not if the hardware level is the one from the leaks.

2- Android Apps? this is the killer feature? Don't we all have a smartphone to use android apps there? +, since when is the NX using android OS? is that a new leak I wasn't aware of?

3- sure, they will probably use cards much like the ds and 3ds did, thus being a handheld



Around the Network
Soundwave said:
bonzobanana said:
People seem to be forgetting the ps4 and xbox one are supported by large capacity optical drives and have internal hard drives for good storage. The NX is a tablet with games coming on lower capacity cartridges and no hard drive. The ambition of NX games will be closer to portable games than home consoles for that reason. NX games will not compare to ps4 and xbone for other reasons than just gpu or cpu performance. If Nintendo really do use very large capacity cartridges we will pay dearly for those games and they will be uncompetitive with ps4 and xbone pricing.

The comparison of NX versus PS4 or xbone will be like wii vs 360 and ps3. What few multiformat games the NX gets will be cut down versions.

It might be an impressive and interesting tablet system but will be a poor home console.

Anyway do we really want to go backwards again with Nintendo supporting a weak system that performs below that of established console products with large software libraries and are just being replaced with much stronger home consoles anyway which are massively superior to NX.

It's highly likely the NX will be dead on arrival unless it really has a gimmick that can save it. That gimmick is going to be far more important than Tegra X1 in NX's success.

Try putting your PS4 in your back pack and playing it on the go. 

Or playing Android apps on your PS4. 

There's your "gimmick". 

32GB (and eventually 64GB) is the same size as a Blu-Ray disc too, so I don't think that's that big of a deal. Even the Nvidia Shield microconsole could accept a 2.5 inch HDD inside its casing too. 

That is my point its a powerful portable system but is not competitive with home consoles. Also remember Nintendo expresses cartridge size in bits not bytes so 32 gig will be 4 gigabytes of memory and 64 will be 8 gigabytes of memory. 



bonzobanana said:
Soundwave said:

Try putting your PS4 in your back pack and playing it on the go. 

Or playing Android apps on your PS4. 

There's your "gimmick". 

32GB (and eventually 64GB) is the same size as a Blu-Ray disc too, so I don't think that's that big of a deal. Even the Nvidia Shield microconsole could accept a 2.5 inch HDD inside its casing too. 

That is my point its a powerful portable system but is not competitive with home consoles. Also remember Nintendo expresses cartridge size in bits not bytes so 32 gig will be 4 gigabytes of memory and 64 will be 8 gigabytes of memory. 

Lets wait to see the specs, it could run even XB1/PS4 games portably, just at a lower resolution.

960x540 is 1/4 the pixels of a 1920x1080 frame, a 400-500 GFLOP Nvidia GPU could certainly run PS4/XB1 ports on the go. The home dock could then run those games at 1080P. That to me makes a lot of sense. 

Dragon Quest XI and Zelda: BoTW both are announced for the system, both are huge games. 

You're wrong about the cart size, 32GB is 32GB, Nintendo hasn't used that metric since like the 1990s, there are already 3DS games larger that are like 4GB. 32GB cards are dirt cheap these days, wouldn't surprise me if Nintendo is able to get those for as cheap as $1 a pop. 



It's an interesting comparison, should also be taken into account the possible usage of the X2 chip as well, which could further improve its performance. It only makes the wait for its official announcement that much more painful T-T



 

NNID: b00moscone

Switch ID: SW-5475-6755-1986

3DS friend-Code: 4613-6380-5406

PSN: b00mosconi

fleischr said:

Even though there's lot's of benchmarks proving the OP correct ... everyone else posting here be like

I thought people who responded made a strong argument. They are saying OP is wrong because simply comparing two PC GPUs cannot show the relative performance of two consoles using similar GPU architecture. a few of the reasons given:

1. the GPU OP used for AMD is old and with old drivers. Driver support is a big factor on PC. Nvidia also has better driver support overall.

2. the AMD GPU he picked is GCN 1st gen while PS4 is similar to GCN 2nd gen.

3. console "code to the metal" meaning they get noticiably more performance out of less power cause the hardware is used more efficiently on a closed platform. DX12 is meant to get more performance out of hardware like a console does. AMD GCN 2.0 also sees a larger DX12 gain than maxwell iirc. with this we can assume that the amd/nvidia flop per flop difference would greatly diminish if the GPUs are used more efficiently as in a console setting (or already seen in DX12 pc games)

4. OP completely ignores power restraint of it being a handheld. Will a handheld be able to utilize a 500 Gflop GPU as well as a console would if it has less power supply and more restricted cooling? OP didn't address this.

 

OP pretty much did exactly the thing he said people shouldn't do; compare GPU by relying too much on flops.

Soundwave said:
bonzobanana said:

that is my point its a powerful portable system but is not competitive with home consoles. Also remember Nintendo expresses cartridge size in bits not bytes so 32 gig will be 4 gigabytes of memory and 64 will be 8 gigabytes of memory. 

Lets wait to see the specs, it could run even XB1/PS4 games portably, just at a lower resolution.

960x540 is 1/4 the pixels of a 1920x1080 frame, a 400-500 GFLOP Nvidia GPU could certainly run PS4/XB1 ports on the go. The home dock could then run those games at 1080P. That to me makes a lot of sense. 

Dragon Quest XI and Zelda: BoTW both are announced for the system, both are huge games. 

You're wrong about the cart size, 32GB is 32GB, Nintendo hasn't used that metric since like the 1990s, there are already 3DS games larger that are like 4GB. 32GB cards are dirt cheap these days, wouldn't surprise me if Nintendo is able to get those for as cheap as $1 a pop. 

pixel count doesn't scale perfectly with GPU power. rendering pixels is only one thing a GPU does. just cause you lower the resolution that doesn't mean the GPU has to render any less polygons or certain graphical effects. Ram size and bandwidth are also important to consider.



Around the Network
FunFan said:

The point is comparing architectures as they are implemented on current gen consoles. Neither the PS4 nor Xbox one use a r7 370. The closest cards to them are the HD 7000 series.

The majority of the 300 series cards were rebadged pieces of hardware from the 200 series. Which were also mostly rebadged parts from the 7000 series.

The Radeon R7 370 is a rebadged Radeon R7 265 which in turn is a rebadged Radeon 7850... Which is almost identical to the Playstation 4 in capability.

torok said:
You can't use the 950 as comparison because it is a 90 watts GPU. A large portable device, such as a tablet, can use 10 to 12 w top. So NX will probably be way behind X1, unless it has some extra stuff when docked.

Agreed.
It's silly using the Geforce 950 as some sort of gauge of Tegra's capability... For one it also has more bandwidth.

JRPGfan said:

^ that is kinda mis leading though.

PS4 & XB1 make use of code to metal.... you want a compairison you should be useing benchmarks that are also close to metal APIs.

Not really. Might be true for a few first party titles, but most games use the low level or high level API's these days due to ease of development.

FunFan said:

Nvidia cards also increase performance with drivers updates as time goes on and the GTX 950 is a newer card, compared to the matured HD 7850 and drivers.

Anandtech is the one I had on hand and I think is trusthworthy.

The "close to the metal" argument is inconsecuential because the NX will also be "close to the metal". This is not a PC vs Console comparison. But a flop per flop performance analisis with the limited resources we have.

Anandtech is very trustworthy... But they are also very slow.
When push comes to shove I will always place Anandtech above all other sites.

On the flip side though, they don't tend to update benchmarks or compare drivers very often like say.. Toms do.

AMD did a ton of work on their drivers fairly recently, overhauling them during their crusades to fix the frame pacing issues... The 7000 series saw some pretty massive gains at that point in time and the older Anandtech benches don't seem to reflect that.

JRPGfan said:

No its not, because they dont "scale" or perform equally in close to metal benchmarks.

AMD does much better in DX12 & Vulkan, than it does in DX11 games (that usually favor nvidia), ontop of the fact that your wrong about the performance increase via drivers for nvidia vs amd. AMD usually scale higher, they launch with terrible drivers, that overtime improve more than nvidia's.

 

If your trying to get a idea, of how a 512 Gflop Tegra X1 would compair to a AMD card, you should be useing DX12 + Vulkan benchmarks.

It's mostly because of Asynchronous Compute that AMD does so well in Direct X 12 and Vulkan.

A 512Gflop Tegra cannot be compared to any desktop chip, plain and simple.
There is a completely different software and hardware ecosystem... Tegra is a "relative" of Desktop GPU's on the PC, but it's still far from the same, it has a completely different memory hierachy for one.

czecherychestnut said:

What you have proved is that flops is  silly metric to use to compare performance with unless you are  only doing compute.  But that's all you can really conclude from your data,  nothing more. 

Even for compute it's a stupid number. Not all compute scenario's use FP32.

HoloDust said:

I don't know why you think that GPUs inside PS4 and XOne are GCN 1st gen. PS4's GPU (due to number of ACEs) looks like 290X/390X cut in half with 2CUs disabled and XOnes is like 7790/260X with 2 CUs disabled (or 360)...they are all 2nd gen GCN.

So you already have pretty good benchmark for 2nd Gen GCN vs Maxwell - 390X vs 980Ti:

http://www.anandtech.com/bench/product/1746?vs=1715

DX12 are Ashes and Hitman.

Now, it is hardly even debatable that nVidia generally has more efficient architectures (looking at 1070@150W vs 480@150W is pretty much all that is needed), it's just that difference when it's DX12 or Vulkan (which is best we can hope for to see in terms of benchmarks) is not as big as DX11 tests where nVidia usually smokes AMD.

The PS4 and Xbox One are GCN 1st gen "with a twist".  -They have SOME features of the GCN 1.1/Gen 2.

A desktop Maxwell is not the same as mobile Tegra. There are similarities for sure, but there are more differences.

nVidia's main edge could be completely due to nVidia's tiled based rasterization, more info on that needs to be found first.


****************

I still believe that in real-world gaming, Tegra will likely be half the performance of the Xbox One in a best-case scenario, provided resolutions are kept low so as not to hit the bandwidth/fillrate limitations of the chip.



--::{PC Gaming Master Race}::--

Good points, on PC's with equal parts the Maxwell architecture design is certainly a lot more efficient.

The issue I see is that the NX will have many bottlenecks holding the GPU back and will limit the efficiency over PS4/X1 by quite a bit. Due to the nature of handheld devices the best RAM we can hope for is LPDDR4, which with a 64-bit BUS at 1600MHz tops out at 25.6 GB/s, meanwhile the PS4 tops out at 176.0 GB/s and the Xbox One at 68.2 GB/s (plus a small pool of eSRAM which has a total of 204 GB/s). They also have 8GB, the NX will likely have under this.

The CPU used in the Tegra is also weaker. Per core the A57 at reasonable clock speeds it beats the Jaguar cores, but the Tegra only has 4 of those vs. 8 in the PS4/Xbox One. It does also have 4 weaker highly power efficient cores (A53) as well, but these aren't used at the same time as the A57 cores used so they don't change power levels at all. Instead they take over for the big cores when the system isn't being taxed much. It's called the big.LITTLE set up and really helps with battery life (which is great).



Bets:

1. If the Wii U sells closer to 10 million LTD by 1/3/2015 I win. If it sells closer to 9.5 million LTD by 1/3/2015 OfficerRaichu15 wins (winner gets 2 weeks of avatar control)--Lost.

Soundwave said:
bonzobanana said:

That is my point its a powerful portable system but is not competitive with home consoles. Also remember Nintendo expresses cartridge size in bits not bytes so 32 gig will be 4 gigabytes of memory and 64 will be 8 gigabytes of memory. 

Lets wait to see the specs, it could run even XB1/PS4 games portably, just at a lower resolution.

960x540 is 1/4 the pixels of a 1920x1080 frame, a 400-500 GFLOP Nvidia GPU could certainly run PS4/XB1 ports on the go. The home dock could then run those games at 1080P. That to me makes a lot of sense. 

Dragon Quest XI and Zelda: BoTW both are announced for the system, both are huge games. 

You're wrong about the cart size, 32GB is 32GB, Nintendo hasn't used that metric since like the 1990s, there are already 3DS games larger that are like 4GB. 32GB cards are dirt cheap these days, wouldn't surprise me if Nintendo is able to get those for as cheap as $1 a pop. 

I'm fairly sure I'm right about cartridge sizes. Normally when Nintendo announces a portable console they give cartridge size information in gigabits not bytes and early 3DS games can be as small as 128 megabytes looking at rom sites. Mario 3D was  512 megabytes but even if wrong there has been no dramatic change between generations in cart sizes, its more evolution and slight cartridge size increase rather than radical change. I.e. if 3DS was  small capacity 128 megabytes, middle 512 megabytes and large 2 gigabytes then probably the next generation will come in at 512 megabytes, 2 gigabytes and 4 or 8 gigabytes top capacity and that would work with the claimed capacities in bits converted to bytes. I dont' think this is surprising as Nintendo still has to pay market value by memory capacity for cartridges.  Its not going to be economic to make true 32 gigabyte or 64 gigabyte games. There is no way Nintendo can compete with the capacity of games for ps4, pc or xbone. Also with cartoon graphics like Zelda and Dragon's quest you don't have the need for such capacity. If the game isn't attempting to look realistic requiring huge texture resources then clearly the game code can be much smaller. 

I'm not interested in buying into a system with cutdown ps4 and xbone games at lower resolution just to take them on the move because portable gaming isn't important to me. There may be others that see that as a huge selling point and I hope they do but I'm not interested in that.

The only reason I'll buy an NX if its as the eurogamer information has stated is NX exclusive games that I feel I must play. However I've come to realise more recently that I'm buying more games than I have time to play and must cut back so they must be pretty much amazing. Zelda U looks like a game like that but I'll use my wii u for that one.  Nintendo are doing a lot of re-releasing of games that I don't need so really it would need to be another huge mario or zelda game. The NX may get a new Zelda game many years in so won't have to buy a NX for that for many years. A new Mario game might temp me but Mario 3D wasn't for me the same quality as something like Mario 64. I think there is a strong possibility that the NX will be far less appealling to me than even wii u. Even now I find both xbox 360 and ps3 both technically  and in the software range superior to wii u. 

Anyway I think people need to have realistic expectations of what the NX will be. Even if the NX has the latest Nvidia chipset its unlikely to be full speed and game cartridges will limit its scope. It's based on hardware more common to android devices than mainstream consoles and is more likely to get enhanced  android ports rather than cut down ps4/xbone ports. It would be huge work to take a top ps4 title and cut it down for NX and the NX may be a more niche console anyway that would not merit it. At least the wii u was a very similar console to ps3 and xbox 360 and so porting from those was easy except for the weaker cpu and lack of hard drive for some games. The NX architecture only enables easy porting of android code. It's highly likely Nintendo are trying to benefit from a fixed platform that can get enhanced versions of android games easily. That's where NX multi-format support is likely to come from. Don't get me wrong these android ports may be amazing and the ultimate versions of such games but pushing the boundaries of games will be done elsewhere.



fleischr said:

Even though there's lot's of benchmarks proving the OP correct ... everyone else posting here be like

Hes skewing the results in favor of nvidia, to make it seem like a 512 Gflop Tegra X1 would be better than it is, compaired to the XB1 & PS4.

The differnce is going to be much less than hes showing, in Tegra X1 Gflop vs XB1/PS4 Gflops.

Im guessing it wont be more than 10-15% better Gflop to Gflop.

A tegra X1 NX would still only be HALF as powerfull as a Xbox One (normal version).

(even with the Gflop to Gflop advantage of nvidias gpus)



NX is not going to compete graphically with Xone or PS4. Why you ask well the most obvious bottlenecks are the powersupply, heat and battery life even if you would supply a gtx1080 with only 20 watts it probably couldn't beat the PS4. And the hybrid NX considering rumours are true also is in need of a CPU which also needs power and generates heat.

 

Even if Nintendo develops thw most efficient architecture ever for the NX (which would probably result in development hell like the cell). It just isn't yet possible for a portable device to come close to Xone and PS4. Especially if you want to sell the hybrid console for less than $300.I suspect the NX will be close to Wii U in therms of power, perhaps a bit more powerful. Graphical cards don't perform equally on PC and a console due to drivers and a more dedicated



Please excuse my (probally) poor grammar