• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

xfire msi 290 oc with 8350

Two completely stock 290Xs @1000/1250

4930k @4.8

Is there any chance of someone with a 8350 and a couple of 290Xs doing a run to compare.


git0.jpg
 
Two completely stock 290Xs @1000/1250

4930k @4.8

Is there any chance of someone with a 8350 and a couple of 290Xs doing a run to compare.


git0.jpg

Wow thats a good score. I think my cards are getting bottlenecked a bit in this game at pci-e 2.0 x8 x8. I've heard that with the new XDMA crossfire implementation there can be bottlenecking on pci-e 2.0 when there are a lot of compute effects at work. I think my max overclocked score was 138 fps Kaap, you're not far off that at stock.
 
I have read this thread with interest (bored at work) and quite shocked at the denial in truth. Sure, a 8350 with a 290 is fine but grab a pair of them and you will be bottlenecked. Run Hitman Absolution/Tomb Raider/Metro LL in game bench tests for evidence.

I am bottlenecked in games if I run my SLI Titans with my 3930K at stock speeds and need to run my CPUat 4.6 to remove that bottleneck. Now I am far from saying get a 3930K but merely pointing out what is what. I see no AMD bashing in this thread however, there is some good home truths which seem to sit uncomfortable with some.

For the OP, he can see that he will be bottlenecked but it is up to him if he decides to grab another 290 and accept that.
 
Athlon 64 was great unless you got caught out buying the single channelled 754(?) socket variant before they released 939 as mainstream just a month or so later...lol.

I had the FX-55 which absolutely destroyed Intel's £800 Prescott CPUs. Ran hot as hell though.

Oh, and the final triumph was the X2 range, which someone in the Electronic Buyer warehouse mistook as two items, so I ended up with a free 4400.

:D.

On topic, it's pretty obvious you'll get bottlenecked with the CPU in question. As Greg says you can become CPU bottlnecked with a hex core i7 quite easily in places with more than one GPU.


Wow thats a good score. I think my cards are getting bottlenecked a bit in this game at pci-e 2.0 x8 x8. I've heard that with the new XDMA crossfire implementation there can be bottlenecking on pci-e 2.0 when there are a lot of compute effects at work. I think my max overclocked score was 138 fps Kaap, you're not far off that at stock.

Kaaps 290X score is monstrous stock and overclocked. Two Ti's @ 1255/3850 and 4960 @ 4.7. That was a while ago though so might give it another run with newer drivers.

j8pi.jpg
 
Last edited:
4770k and 2x 290ps, money saved getting P instead of X might just be enough to get a 4770k. You'll get better FPS eitherway and a more rounded system.

Edit:

YOUR BASKET
2 x MSI Radeon R9 290 Gaming Edition 4096MB GDDR5 PCI-Express Graphics Card £359.99 (£719.98)
1 x Intel Core i7-4770K 3.50GHz (Haswell) Socket LGA1150 Processor - Retail £239.99
1 x MSI Z87-G43 Gaming Series Intel Z87 (Socket 1150) DDR3 ATX Motherboard £109.99
Total : £1,079.56 (includes shipping : £8.00).




YOUR BASKET
2 x MSI Radeon R9 290X Gaming Edition 4096MB GDDR5 PCI-Express Graphics Card £449.99 (£899.98)
1 x AMD Piledriver FX-8 Eight Core 8350 Black Edition 4.00GHz (Socket AM3+) Processor - Retail £149.99
1 x Asus Sabertooth 990FX R2.0 AMD 990FX (Socket AM3+) DDR3 Motherboard £139.99
Total : £1,199.56 (includes shipping : £8.00).


Why the more expensive motherboard for the AMD setup?? Any of the Asus EVO motherboards have a solid VRM section which can push upto 275W alone. You also bumped up the price more by adding R9 290X cards to the AMD setup instead of R9 290 cards for the Intel setup and on an overclocking motherboard,an FX8320 would be fine,saving you upto £40 over an FX8350 - OcUK had them for under £110 on offer recently.

That would make the AMD setup £200 to £250 cheaper,with an FX8320,990X motherboard and R9 290 cards.

The R9 290X and R9 290 are so close together,the R9 290X is not really worth the extra cash anyway. Also as the OP indicated in his posts,a single R9 290/R9 290X should be enough for the games he plays.
 
Last edited:
Way to quote out of context, bro...
Check who I directed that at, somebody with a 290x, 8350 and that mobo. My point is he could have got a much better more rounded system for less money buying the cheaper Pro cards. If you've got a problem mate direct it at pando not the OP.
 
the games i am playing are ...
war thunder
battlefield 4 / i have tried mantle driver and it increases my fps and performance
thief when its out
titanfall when its out
company of heroes 2
cs go
arma 3 + altis life/breaking point
loudout
diablo 3 and expansion when its out

Lets have a look at your games OP.

War Thunder is CPU light:

http://gamegpu.ru/images/remote/htt...est_GPU-MMO-War_Thunder_1.37-test-wt_proz.jpg

BF4 has Mantle updates and even before then,Sweclocker tested Conquest maps with a R9 290X:

http://www.sweclockers.com/image/diagram/4970?k=31c697c537d6c542a5c7cc66fb1aa243

So its around 10% slower than a Core i5 4670K on Ultra. Sweclockers tested with 50+ people on each server and multiple 3 minute runs. One or two other sites used smaller domination maps with lower player counts to test MP and these were not very valid. On medium settings the Core i5 does push a bit ahead,but I suspect you are running on higher settings anyway. However,with Mantle framerates and overall frame latencies should improve especially during MP scenarios,so I personally think your current overclocked R9 290 should be fine.

Thief is a UE3 game and includes Mantle support. Interestingly it means Epic will probably have Mantle support on UE4 if they enabled it in UE3. So that is probably UE4 to the list plus CryENGINE 3 and Frostbite 3.

Since it is a UE3 based game it should run fine especially with Mantle support.

Titanfall uses the Source engine. I expect that it should be fine on your setup - source based games run fine on most hardware. It has modest minimum hardware specs:

http://www.pcgamer.com/2014/02/04/titanfall-minimum-system-requirements-revealed/

CS:GO runs fine on old hardware and is another Source based game.

Loadout is F2P,so it won't be too taxing on hardware as it would scare people away.

Diablo3 will run fine on your setup - I have mates who have worse hardware than you who play it at 1920X1080 at max settings fine.

There are two benchmarks for CoH2:

http://gamegpu.ru/images/stories/Test_GPU/strategy/Company of Heroes 2-2/test/coh proz.png
http://static.techspot.com/articles-info/689/bench/CPU_01.png

One places the FX8350 around 30% slower than a Intel Core i5 or Core i7 and another one around 10% slower.

ARMA3,and indeed the whole ARMA series seems to do better on Intel CPUs,so it is a moot point!!:p

So,OK for two of the games,it seems an Intel CPU would be faster,but the rest will run more than well enough.

However,since CoH2 is not an FPS game,IMHO,it should be OK with slower or older CPUs.

ARMA3 is probably the main problem IMHO.

Way to quote out of context, bro...
Check who I directed that at, somebody with a 290x, 8350 and that mobo. My point is he could have got a much better more rounded system for less money buying the cheaper Pro cards. If you've got a problem mate direct it at pando not the OP.

Ahh,serves me right for skimming the latter part of the thread.

I agree with what you were saying.

However,sadly the OP did not spend the extra £10 to £15 on M5A97 EVO R2.0 instead of the M5A97 R2.0.
 
Last edited:
Pando is a new one, I like it :)

Agreed that the pro is much better price/performance than the X. I think the point I was making was lost in the exchange tho.

There will always be a cpu bottleneck in certain games with any setup. I think the bottleneck aspect is played up to be more of a big deal than it is though. Oft times you will be better off spending more on the gpu grunt than the extra for an intlel setup that alleviates the bottleneck.
 
No offence, but to me, it seems that mainly people who have bottlenecks who seem to downplay the importance of them, at least from personal observation.

When I switched to Intel it's because I had no other path, AMD killed off AM3, and my Thuban was already bottlenecking a 6870 Crossfire in titles at that time (Mainly DOW II/SWTOR as I played both heavily)
Sure, I could have thrown more grunt at it, and see gains, but it overall just felt much nicer without the bottleneck, smoother etc.
 
Last edited:
The vast majority of gamers will be bottlenecked in some way,ie,either CPU or most likely GPU. There is always new hardware being released,and hence you will always potentially get higher framerates with the latest and greatest.
In the end as long as your are getting enough performance it does not really matter IMHO otherwise you will go neurotic with benching 24/7 and never being happy,and always buying new hardware. When I was younger I went through a phase of buying loads of computer hardware,benchmarking,etc and now I look back and realised I ****ed away money for nothing - I could have done more useful things with it TBH.

OTH,the marketing of large tech companies exploits the fact- why have that 1 year old phone when the latest one is 25% faster,etc. Most people would not tell the difference if both phones looked the same.

Like I said great if buying new hardware is your hobby, but then it makes me wonder at times if a double blind study was done whether people would notice different setups or settings that easily or whether it is confirmation bias.
 
Last edited:
i got a [email protected] and it does bottleneck even 1 R9-290X.Most of the time in games which not using all cores(so most of them).IPC of amd chips sucks.Get a 4770K and done with it.Im going to do the same too.

I setup a friends 8350 @ 4.6 with a 290x and it was a bottleneck with just one graphics card. He wanted 8 cores anyway.
 
Aema 3 is a big issue. i got terrible performance with arma , 3 and even dayz stand alone. The dips in it made shooting accurately when zombies were around, impossible. not nice when the game is permadeath.
 
The vast majority of gamers will be bottlenecked in some way,ie,either CPU or most likely GPU. There is always new hardware being released,and hence you will always potentially get higher framerates with the latest and greatest.
In the end as long as your are getting enough performance it does not really matter IMHO otherwise you will go neurotic with benching 24/7 and never being happy,and always buying new hardware. When I was younger I went through a phase of buying loads of computer hardware,benchmarking,etc and now I look back and realised I ****ed away money for nothing - I could have done more useful things with it TBH.

OTH,the marketing of large tech companies exploits the fact- why have that 1 year old phone when the latest one is 25% faster,etc. Most people would not tell the difference if both phones looked the same.

Like I said great if buying new hardware is your hobby, but then it makes me wonder at times if a double blind study was done whether people would notice different setups or settings that easily or whether it is confirmation bias.

Well it depends really. I know for me personally, it makes no sense to have GPUs bottlenecked by a CPU. Of course the bottleneck has to be somewhere but if you've got a load of graphics hardware not running at full throttle then that isn't ideal.

All you can ask is that if you've bought hardware then it's running to its realisable potential. Having a CPU bottleneck is not allowing that.
 
thanx for that cat, yes guys i know the intel i5/i7 is better for gaming generally but at the time i couldnt afford it and i didnt realise at the time how sucky the m5a97 r2 is at overclocking. yes i should have bought one of the better ones

i wont be going xfire for the moment. i cant afford a new 290 a new mobo AND an intel cpu
 
No offence, but to me, it seems that mainly people who have bottlenecks who seem to downplay the importance of them, at least from personal observation.

When I switched to Intel it's because I had no other path, AMD killed off AM3, and my Thuban was already bottlenecking a 6870 Crossfire in titles at that time (Mainly DOW II/SWTOR as I played both heavily)
Sure, I could have thrown more grunt at it, and see gains, but it overall just felt much nicer without the bottleneck, smoother etc.


None taken. Imo, the word Bottleneck is over emphasized to spread fud sometimes in this forum.

Yeah its good advice to say there is a bottleneck and performance would be increased with a platform change. But that isn't necessarily the best way to spend yor money if on a budget.
 
Back
Top Bottom