384bit vs 512bit memory bus Tomb Raider shootout

[Kaapstad]

Just for a bit of fun I have done some tests using the Tomb Raider benchmark to compare the memory buses on the Titan (384bit) against the R290X (512bit).

The PCs used were near identical with

Mobo = RIVE
Ram Corsair Dominator 2400mhz 9-11-11-25
CPUs 3970X and 3930k both @4.0ghz

To test what I did was

First set the VRAM on both cards to 1251 (this is the lowest I could set on the Titan).

Next I ran the TR bench @2560 x 1600 (I used this resolution to max the VRAM workload) on both the PCs and overclocked the Titan on the GPU core until both the Titan and 290X were scoring the same (25.4fps).

For the actual tests the only changes I made was to increase the VRAM speed on both cards by the same amount to compare what increase in fps I got.

The reason I did this was in theory the 512bit bus should give more bandwidth and performance for any given increase in VRAM clock.

R9 290X starting settings

Titan starting settings

Results with the VRAM @1251mhz (starting point)

290X

Titan

VRAM now at 1350mhz

290X

Titan

VRAM now at 1450mhz

290X

Titan

VRAM now at 1550mhz

290X

Titan

VRAM now at 1600mhz

290X

Titan

Please remember this was just one game and one resolution.

Below is a summary of the results in the pics


VRAM @1251mhz 290X 25.4, Titan 25.4

VRAM @1350mhz 290X 26.2, Titan 26.1

VRAM @1450mhz 290X 26.5, Titan 26.8

VRAM @1550mhz 290X 26.8, Titan 27.0

VRAM @1600mhz 290X 26.9, Titan 27.2


One possible reason for the Titan pulling ahead is that error correction could have come into play with the 290X.

 

[Kaapstad]

Well this game doesn't seem to respond well to memory clock increases.

Thanks Kaap, interesting to see.

The increases look small because of the resolution used but going from 25.4 to 27.2 is nearly 7.1% increase in fps.

I wanted to use 1600p to give the memory buses some real work to do.

 

[Kaapstad]

Interesting Kaap, thanks for posting.

Pretty close. I wonder if there is a tipping point at which a higher memory bus provides benefit over a slower bus with more VRam and at what res that might be.

I think you would need to look at 4K and above to go with the slower vram wider bus solution. But that is only my guess, like all things it needs testing.

 

[Kaapstad]

He does not start of at default on the titan though. He starts at 1250 the same as the 290 but at 1250 on the titan thats only 240 gb/s compared to 320 gb/s on the 290.

I just had a quick go on the Heaven 4 bench @1600p as well but this time it was the 290X that had to be overclocked on the core to match the Titan at stock. Again I tested the VRAM @1251mhz and 1600mhz. The performance increase by raising the VRAM speed on both cards was within a single point of each other.

The only thing I can think is the width of the bus is playing no part at this resolution (1600p) on Heaven 4, only raw VRAM mhz.

 

[Kaapstad]

Kaap can you do the same bench with the two cards again using the same settings and methods but with 0xAA, so that it can be compared against the 4xSSAA results?

The reason for this is to see if higher memory bandwidth does help reduce the amount of frame rate drop in % when applying higher level AA. If the Titan has quite a bit higher frame rate on 0xAA, that would mean higher memory (on the 290x) bandwidth does reduce the amount of performance hit.

With 0xAA the Titan does run faster than a 290X but once I underclocked the Titan slightly (-13mhz on the core to get both cards to the same starting point) the effects of changing the VRAM speed were much the same as before.

This is what I got with no AA

VRAM @1251mhz (starting point) I settled for getting the two cards within 0.1fps of each other.

290X

Titan

VRAM @1350mhz

290X

Titan

VRAM @1450mhz

290X

Titan

VRAM @1500mhz

290X

Titan

1500mhz was as high as I could go on the 290X, I don't think the memory liked the higher fps without AA.

As I said earlier I had to underclock the Titan slightly so both cards were starting from the same point @1251mhz on the VRAM, but even with the Titan overclocked on the core the increase in fps when changing the VRAM settings was not much different to what I have posted above with the cards starting from the same point.

The VRAM speed changes seems to behave the same whether the cards are using AA or with it disabled and also when the Titan is overclocked or underclocked on the core to reach the same starting point. I suspect that neither card is using it's full bus width and the results are based mostly on VRAM speed changes.

 

[Kaapstad]

So when partner companies move to fast vram modules and better coolers on the 290/x, will these cards be much faster than today using the same gpu core?

At normal resolutions I think the faster ram will give higher fps but at higher resolutions like 4K a wider bus will be more beneficial.

So my guess is yes the cards will be about 5% faster.

 

[Kaapstad]

From stock (1250) or 1500? from 1250 to 1750 i would say its a bit more than 5%, its a 40% speed increase, that should yield 15 to 20%, from 1500 to 1750 about 7 to 10%.

Using TR in the OP as a guide going from 1251mhz to 1600mhz gave about 6% increase. Having said that it will vary from game to game, it all depends whether the VRAM is causing a bottleneck in the first place.

 

[Kaapstad]

I would test it myself but have unistalled it to make room for other games. i don't think TR is much dependant on memory speed, its far more GPU heavy, your memory also got to its limits past 1500Mhz, giving you no gain.

If i remember rightly BF3 was almost equal in Memory percentage gain to GPU.

Out of interest what memory IC's are on it?

Just ran this on the Titan

The first run is one I did earlier with the core underclocked by -13mhz and the VRAM @1251mhz

The second run is what I have just done with the core at the same speed (-13mhz) but the VRAM @1902mhz

Going from 1251mhz to 1902mhz on the VRAM

The increase is 4.8 fps or 10.19%

I know there will be games that show a bigger improvement when overclocking VRAM but there will also be games that don't.

As to what chips I have on my 290Xs, I have never bothered to check as they are all good clockers on the VRAM. They are all Asus 290Xs so whatever they come with.

 

[Kaapstad]

Yes....?

The point is that the Titan isn't gaining more proportionately than the 290X so it looks like it isn't bandwidth limited in this game. It doesn't matter how much GB/s it is if it isn't limited on this front. The gains are linear suggesting improvements from the speed alone rather than bandwidth.

+1

It does not seem to matter if the Titan is over or underclocked on the core, when changing the VRAM speed the gains are linear.

 

[Kaapstad]

You need 8Ghz on a 512-bit bus . That'll learn Nvidia.

I don't think it will be long before AMD or NVidia do it if they are serious about 4K.