• 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.

intel i5's vs amd FXs

I game on one of my AMD systems which is an 8320 @ 5.1 with my 290x at 1241/1500 and even when i had the CPU at 4.4, the only bottlenecks i saw was occasionally in old RTS games but even then, i could only notice with fraps and GPU-z open. No one doubts that they cant bottleneck but it doesnt really matter if it only really bottlenecks in old games which have playable/good frames anyway.
 
The GPU usage monitoring in the R9 290/290X is broken, has been since launch.

A 4.4GHZ FX8350 would bottleneck an R9 290 in a load of games.
My R9 290 at 1150MHZ outmatches a 7990 with FX83 in the Resi benchmark, that's the only time I've ever went to do a bench like that, but I didn't post the result, no point.

It was my i5 4670K at stock, and Alex was running his FX8350 overclocked.
 
Last edited:
Martini1991 said:
The GPU usage monitoring in the R9 290/290X is broken, has been since launch.
Click to expand...

I know its broken in a way that the load would report as dropping randomly to 0, but clearly is not doing that during games or mining. I cant tell for sure if it is being bottlenecked in these old RTS's but either way, i am still getting decent frame rates in everything i have played except for the alpha Dayz Standalone which was released last week (putting that one down to poor coding like the original mod, glitchier than BF4 imo!).
 
Martini1991 said:
There's a few games that give less than desirable performance.
If you were using 3D via Tridef, the AMD CPU's would be giving you less than desirable performance too.

But if we're using the "Do you need anymore" logic, we wouldn't be running the GPU set ups that we do.

Since most people don't use the type of set up I use (Or other people) AMD CPU's would be fine, it's for those higher end ones that they're not fine for (Hence why I only really start pimping out the Intel chips in budgets that can support it)
Click to expand...

In a situation where a properly overclocked FX 8-core is giving you less than desirable performance in tridef 3D (which I regard as sub 30 fps), the i5 4440 would almost certainly be causing the same issue and you'd have nowhere to go in terms of overclocking that.

Martini1991 said:
The GPU usage monitoring in the R9 290/290X is broken, has been since launch.

A 4.4GHZ FX8350 would bottleneck an R9 290 in a load of games.
My R9 290 at 1150MHZ outmatches a 7990 with FX83 in the Resi benchmark, that's the only time I've ever went to do a bench like that, but I didn't post the result, no point.

It was my i5 4670K at stock, and Alex was running his FX8350 overclocked.
Click to expand...

Not sure if you meant me, but I did run the resident evil benchmark. Haven't run it with my 780 to compare. From memory, it used 1/2 cores/threads and even then, pretty poorly. I wouldn't use that as a representative measure of FX 8-cores bottlenecking single GPU setups though. It obviously isn't a game and we have no idea of what the total score is composed of as the benchmark doesn't give details of CPU/physics scores vs GPU/graphics score.
 
alex_123_fra said:
In a situation where a properly overclocked FX 8-core is giving you less than desirable performance in tridef 3D (which I regard as sub 30 fps), the i5 4440 would almost certainly be causing the same issue and you'd have nowhere to go in terms of overclocking that.
Click to expand...

Are you assuming there? Because I don't know either way how a locked i5 would handle it? I know my clocked i5 4670K was perfect, and I know clocked FX83's aren't.

alex_123_fra said:
Not sure if you meant me, but I did run the resident evil benchmark. Haven't run it with my 780 to compare. From memory, it used 1/2 cores/threads and even then, pretty poorly. I wouldn't use that as a representative measure of FX 8-cores bottlenecking single GPU setups though. It obviously isn't a game and we have no idea of what the total score is composed of as the benchmark doesn't give details of CPU/physics scores vs GPU/graphics score.
Click to expand...

Yeah, it was you.
Resi benchmark is far more representative than BF4 etc is for not bottlenecking (As many people seem to use etc)
 
Last edited:
I do have a 8320 and a 3570k rig.

You guys probably already know which games the 8320 has trouble with.

I think the 8320 is great price at £109 obviously but for gaming I choose intel every time.
 
Martini1991 said:
Are you assuming there? Because I don't know either way how a locked i5 would handle it? I know my clocked i5 4670K was perfect, and I know clocked FX83's aren't.
Click to expand...

Yes I am, but I am not talking about an i5 4670k. I am talking about an i5 4440. Even at stock the k chip has a fairly large 500Mhz advantage in boost clock. Your 4670k was also likely overclocked.
 
Martini1991 said:
Yeah, it was you.
Resi benchmark is far more representative than BF4 etc is for not bottlenecking (As many people seem to use etc)
Click to expand...

Not sure about that. I also seem to remember a discussion with almighty about how there was evidence that the resident game the benchmark was based on, was able to utilise many threads (4+) well. None of us saw that in the bench. It would worry me if a game and it's bench had such a disparity in core/thread utilisation as I won't play the bench, I'll play the game. Again, I also don't know the physics/graphics split of any of the results so ideally to conclude this we would need to run it with the same gpu (at the same clocks).
 
I don't think the benchmark is very threaded, but neither are most games, that's the point I'm making.
Crysis 1 when it came out and for a while showed no difference with CPU, but now you can bottleneck that with an FX83 (Higher end GPU set ups) Or at least according to Almightys results.
 
Martini1991 said:
I don't think the benchmark is very threaded, but neither are most games, that's the point I'm making.
Crysis 1 when it came out and for a while showed no difference with CPU, but now you can bottleneck that with an FX83 (Higher end GPU set ups) Or at least according to Almightys results.
Click to expand...

Sure, but the point I was making was that an i5 4440 won't be in the same league as an overclocked 4670k (or even stock clocks for that matter). When comparing the former to an overclocked fx 8-core, I don't think you'll find a huge difference in terms of bottlenecking single gpu setups in poorly threaded games.
 
alex_123_fra said:
Sure, but the point I was making was that an i5 4440 won't be in the same league as an overclocked 4670k (or even stock clocks for that matter). When comparing the former to an overclocked fx 8-core, I don't think you'll find a huge difference in terms of bottlenecking single gpu setups in poorly threaded games.
Click to expand...

I somewhat agree, hence why I've never really made mention to the locked i5 and why I rarely if ever recommend them, hence my original post saying I don't think they're worth buying.

Forgetting of course that the OP isn't on about the locked i5 :p
 
Martini1991 said:
Don't know what Almighty was running.

Pretty sure I'd encounter a bottleneck with an R9 290 clocked at 1150MHZ though.

Phix was using an FX83 with a 780Ti, and he's posted about it bottlenecking.
Click to expand...

I have only played a few hours of it since grabbing this card and didn't run any monitoring software, so i cant comment on if it bottlenecks but I found it to perform quite well still (a fair boost in performance compare to the 580 i had in this rig before).
 
I wouldn't monitor GPU usage with an R9 290, what Unwinder says is different to what you say from what I recall.

But Almightys benchmark results with the predetermined benchmark showed a bottleneck, not that it's surprising, Crysis uses 2 cores.
 
Martini1991 said:
I somewhat agree, hence why I've never really made mention to the locked i5 and why I rarely if ever recommend them, hence my original post saying I don't think they're worth buying.

Forgetting of course that the OP isn't on about the locked i5 :p
Click to expand...

Agree. The only reason I came back into the thread was following imginy's post (#9) and the specific mention of the i5 4440.
 
intel i5 and amd fx are reasonable cpus for the price

fx 8320 only one worth buying and ocing rest are pointless .

i5 vs fx amd. intels are faster for gaming there is no debate about it.

on a budget you choose amd. for performance you choose intel .this wont change even with mantle. only performance you'll gain is roughly in frostbite games which are literally none except bf. so basically the gap that the fx is down on the i5s it will probably level in bf4 that's it. all rest as usual slower.

amd do offer good and great bang for buck just not to intels standard.
 
alex_123_fra said:
Sure, but the point I was making was that an i5 4440 won't be in the same league as an overclocked 4670k (or even stock clocks for that matter). When comparing the former to an overclocked fx 8-core, I don't think you'll find a huge difference in terms of bottlenecking single gpu setups in poorly threaded games.
Click to expand...

This is so far from the truth.

Look at these benches nearly twice the framerate on a 3.4 haswell vs a 4.8 8350.

http://www.bit-tech.net/hardware/2013/11/14/intel-core-i3-4130-haswell-review/5
 
You must log in or register to reply here.
Back
Top Bottom