Mantle vS 780Ti in BF4

[gpuerrilla]

It seems even with the reported 8% performance gains from using Mantle, it still wasn't enough to beat the 780Ti in their findings on a 4770K CPU.

Doesnt exactly look like the 780ti is being ridiculed

Green detector siren - Woooooot, Wooooooot

They used Medium/High settings.

The question you ask yourself is who spends £550 on a high end gpu to play at lower details?

This ^ . Also the 780ti is £100+ more than the 290x, hardly any bragging rights. Is this for self justification on blowing money or aimed at stirring - I'm not sure.

 

[Silent_Scone]

I'm at work but if I remember I'll add my ten cents. At what CPU clocks shall I run my 4960 though? .

Stock, 4.5, 4.7? Ahhhh too many variables

Green detector siren - Woooooot, Wooooooot



This ^ . Also the 780ti is £100+ more than the 290x, hardly any bragging rights. Is this for self justification on blowing money or aimed at stirring - I'm not sure.

Again, read the purpose of the article before getting your panties all in a bunch


"I STRONGLY disagree with how that chart looks. You spent more money than me so NERGGH"

 

[Gregster]

I am reporting news from a well respected tech site and as I also like nVidia, I thought it was some good news to share with nVidia users. Sorry if you didn't like it but there is plenty of pro AMD threads

It seems fine to praise AMD and yet try doing the same for nVidia users and you get this....

Green detector siren - Woooooot, Wooooooot

Lighten up guys, we all enjoy our gaming I hope and doesn't matter what GPU we have does it? News is news and we don't always like it but suck it up.

Prove them wrong by running that bench. Mantle is supposed to help with the CPU bottleneck Matt, so it should give AMD owners an advantage surely?

 

[LtMatt]

Please feel free to add 25Mhz to the clocks and run again and anyone with a 290X, feel free to join in and likewise with a 780Ti. Remember, stock clocks and full ultra settings.

Frames, Time (ms), Min, Max, Avg
6684, 118389, 42, 74, 56.458

I know yours is a 290 Matt but Mantle isn't looking like it will ridicule a 780/Titan in BF4 with a 290X just yet

Couple of things Gregorio. I had an i7 4 core vs your 6 core. I'm also using a £300 290 which has less Shaders, compute units, TMU's, is clocked slower and is about 10% slower than a 290X. Adding 25mhz onto the core will not make up the difference. I also noticed in your benchmarking run you didn't follow the instructions i posted so not that consistent to compare really. But if it makes you feel better, go for it.

 

[Gregster]

Couple of things Gregorio. I had an i7 4 core vs your 6 core. I'm also using a £300 290 which has less Shaders, compute units, TMU's, is clocked slower and is about 10% slower than a 290X. Adding 25mhz onto the core will not make up the difference. I also noticed in your benchmarking run you didn't follow the instructions i posted so not that consistent to compare really. But if it makes you feel better, go for it.

I am not a robot sorry but tried my best to stick to the way you did it.

Read my post above as well about the bottleneck.

 

[LtMatt]

I am not a robot sorry but tried my best to stick to the way you did it.

Read my post above as well about the bottleneck.

If you want to compare, you need to make it a fair comparison is all im saying. If you want to accurately compare results you need the same cpu and to follow the same procedure. I went out of my way in that bench to shoot all explosive barrel as part of the timed benchmark sequence. Unless you follow that like i did, its not an accurate comparison. I wrote down the instructions on the video of the protocol to follow for consistency. As you don't have an AMD gpu though its largely pointless as that time demo was for the Mantle thread.

 

[uksoldierboy]

Forgive me i don't have the time to crop and crap, really need my full breaky! lol

 

[LtMatt]

So 13% faster than Gregorio's titan. ^

 

[Gregster]

If you want to compare, you need to make it a fair comparison is all im saying. If you want to accurately compare results you need the same cpu and to follow the same procedure. I went out of my way in that bench to shoot all explosive barrel as part of the timed benchmark sequence. Unless you follow that like i did, its not an accurate comparison. I wrote down the instructions on the video of the protocol to follow for consistency.

Give over Matt. I can't do any more without buying a complete new system with the same ram/mobo/psu/cpu as you before it would be considered and then you will pick hairs that I might have done something else different.

I am not trying to convince you or anyone else to buy a 780Ti and was just reporting some news that might have been percieved as good by nVidia owners (including myself). I didn't realise that was so wrong and frankly, I am getting a little wound up.

I have given people the opportunity to copy your bench and post their results (for a bit of fun) but don't bother if you feel it is pointless.

 

[Silent_Scone]

Yeah, see now you're happy, Matt lol ^_^.

You're never going to get clear results without a fixed bench. Not even TR's findings. I'd expect a 290X to be faster than a Titan though with Mantle, most definitely.

But your frame count is a couple thousand less than Greg's test. It's all senseless, just enjoy the smoothness

 

[LtMatt]

Give over Matt. I can't do any more without buying a complete new system with the same ram/mobo/psu/cpu as you before it would be considered and then you will pick hairs that I might have done something else different.

I am not trying to convince you or anyone else to buy a 780Ti and was just reporting some news that might have been percieved as good by nVidia owners (including myself). I didn't realise that was so wrong and frankly, I am getting a little wound up.

I have given people the opportunity to copy your bench and post their results (for a bit of fun) but don't bother if you feel it is pointless.

Relax Gregorio. You don't have to prove anything to me. Just saying if you want an accurate comparison then you'll need similar hardware, particular at a res of 1080P on this scene where the cpu plays a part in the results.

 

[Gregster]

So 13% faster than Gregorio's titan. ^

A very very low frame count on Recce's run (5760).

1080P results
Frames, Time (ms), Min, Max, Avg
7703, 91042, 68, 118, 84.609

You need to crop that image please Recce.

 

[LtMatt]

Yeah, see now you're happy, Matt lol ^_^.

You're never going to get clear results without a fixed bench. Not even TR's findings. I'd expect a 290X to be faster than a Titan though with Mantle, most definitely.

But your frame count is a couple thousand less than Greg's test. It's all senseless, just enjoy the smoothness

Well said.

 

[gpuerrilla]

Again, read the purpose of the article before getting your panties all in a bunch

The facts are facts though Frosty. Mantle is showing signs of what it was intending to do. Telling me to read when people are not reading themselves (i.e. Greg not following a guide created by Matt is ironic) is sidestepping the point I was making.

It seems fine to praise AMD and yet try doing the same for nVidia users and you get this....

Lighten up guys, we all enjoy our gaming I hope and doesn't matter what GPU we have does it? News is news and we don't always like it but suck it up.

There is nothing to suck up. No one is after the praising of AMD, but trying to come off as a neutral when some here are clearly all for nVidia is a bit rich.

Lighten up is right - best thing you have said. Driving down prices of hardware and API's that give free performance boosts is a GOOD thing. Quit with the brand hatred it's so childish.

 

[Nickolp1974]

Not really following any of these mantle threads, mainly due to all the bickering! Always seem to end up green v red argument again.

But to put an end to it here's one suggestion,
This probably goes to Kaap

Clock a Titan to 780Ti levels and say run fire strike, now match a 290x to same gfx score. Now we have comparable cards, yes???
Now test each card at there comparable clocks on bf4, this should give a fairly representative answer to which is better then you can all get back to the real topic in hand.

 

[Gregster]

I give in and that will do.

 

[LtMatt]

Not really following any of these mantle threads, mainly due to all the bickering! Always seem to end up green v red argument again.

But to put an end to it here's one suggestion,
This probably goes to Kaap

Clock a Titan to 780Ti levels and say run fire strike, now match a 290x to same gfx score. Now we have comparable cards, yes???
Now test each card at there comparable clocks on bf4, this should give a fairly representative answer to which is better then you can all get back to the real topic in hand.

That sounds like far too much effort lol.

 

[Suarez7]

780, 780ti, 290, 290x all going to be smooth on Ultra unless you're using a massive res. Why not just enjoy the game if you like playing it instead of trying to willy wave running benches every day or so. I'm past caring which card gets 5 frames more every second now and with a game like BF4 where it can't be benched fairly I think some of you should move on from caring as well.

 

[LtMatt]

780, 780ti, 290, 290x all going to be smooth on Ultra unless you're using a massive res. Why not just enjoy the game if you like playing it instead of trying to willy wave running benches every day or so. I'm past caring which card gets 5 frames more every second now and with a game like BF4 where it can't be benched fairly I think some of you should move on from caring as well.

Also well said.

 

[gpuerrilla]

Not really following any of these mantle threads, mainly due to all the bickering! Always seem to end up green v red argument again.

But to put an end to it here's one suggestion,
This probably goes to Kaap

Clock a Titan to 780Ti levels and say run fire strike, now match a 290x to same gfx score. Now we have comparable cards, yes???
Now test each card at there comparable clocks on bf4, this should give a fairly representative answer to which is better then you can all get back to the real topic in hand.

I dont know if the thread was intending to prove which card is 'better'. I think showing the improvements mantle makes (if any) and allowing it to stretch it's legs (on weaker systems) is the end goal.