The Open-Source Linux Graphics Card Showdown

[AnonymousCoward]

So my ?100 radeon card is more than twice as fast (on the more demanding games) then this ?300 Intel thing.

[Kano]

Well all cards are tested with the same cpu if your driver as high cpu usage then a slow cpu will not give you the same result. In most cases the single core speed of a cpu helps to scale with older engines - if you are searching for benchmark records, over 60 fps is usally never needed. doom 3 is certainly a bit slow, should be compared with win maybe to see the raw performance of the chip. If you like you can oc it a bit as well, with simple engines it should scale well, the only "reference" oc test i found was using bf3. Of course for several cards you can install binary drivers and get much more speed, intel has got only one driver - if the performance is not fast enough for you then all you can do is to add an extra card. For desktop systems thats a piece of cake. But laptops the intel hd 4000 seems to be a good choice, no hybrid solutions which only work with weird hacks.

[Kano]

I don't think that you can run Prey with iodoom3, most likely not even Quake4. You can play just Doom 3 + Res. of Evil.

[Kano]

Its most likely similar but not the same, best ask icculus.

[Drago]

for his question it just doesn't matter.

"similar" means similar but he writes like "there is no similar"

please translate this in your autistic language.

Forum troll strikes again! Prey uses MODIFIED iDTech4. Iodoom3 is not going to work.

[droidhacker]

Well all cards are tested with the same cpu if your driver as high cpu usage then a slow cpu will not give you the same result. In most cases the single core speed of a cpu helps to scale with older engines - if you are searching for benchmark records, over 60 fps is usally never needed. doom 3 is certainly a bit slow, should be compared with win maybe to see the raw performance of the chip. If you like you can oc it a bit as well, with simple engines it should scale well, the only "reference" oc test i found was using bf3. Of course for several cards you can install binary drivers and get much more speed, intel has got only one driver - if the performance is not fast enough for you then all you can do is to add an extra card. For desktop systems thats a piece of cake. But laptops the intel hd 4000 seems to be a good choice, no hybrid solutions which only work with weird hacks.

You bring up an important point. Following that fact, since the INTEL GPU is IGP *in* the CPU, any power consumption comparisons for it are meaningless unless the CPU is swapped over to one withOUT IGP. The thing is going to sit there sucking down at least SOME power, whether you're running it or not.

I also question the performance tests. Intel is known for playing games in order to benchmark higher. It wouldn't surprise me one bit if there was something sneaky in the CPU that knows when someone else's GPU is attached and does something like cutting memory bandwidth. Intel is known for pulling sneaky crap, like designing CPUs FOR the benchmarks, rather than the real workload.

Some people are going to flame me for this, but the reason why Intel beats AMD on benchmarks is because AMD is actually innovative -- reinvent the wheel, then wait for software to catch up, rather than bolting a supercharger onto an '85 Lada and calling it fast.

[Kano]

Thats incorrect. The gpu part has got different voltage pins, you can select the differently to gpu voltag on Z boards. The default is that it is disabled as soon as you add a PCI-E gfx card, but you can force it to stay enabled if you want to use it together with virtu (on win). Intel even sells chips with disabled gpu part, not yet for ivb, but for snb like:

http://ark.intel.com/compare/52273,52274

Just a matter of time, then intel has got too many ivb cpus without working gpu and wants to sell those too

[Nille_kungen]

Nice test, but I think it's near impossible to tell which curve corresponds to which graphics card in those graphs. The colors are way too similar.

Thank you i thought it was only me.
There are so many colors so why only use shades of the same colors.
Why not use distinct colors instead?
Example: brown, black, red, orange, violet, blue, green, yellow, turquoise, pink.

[droidhacker]

Thats incorrect. The gpu part has got different voltage pins, you can select the differently to gpu voltag on Z boards. The default is that it is disabled as soon as you add a PCI-E gfx card, but you can force it to stay enabled if you want to use it together with virtu (on win). Intel even sells chips with disabled gpu part, not yet for ivb, but for snb like:

http://ark.intel.com/compare/52273,52274

Just a matter of time, then intel has got too many ivb cpus without working gpu and they want to sell those too

Its not just about turning the part off. Yes, it will use less, but we're talking about semiconductors here, not a mechanical circuit breaker. Go back to your hardware engineering classes and remember that 0v is not equal to exactly 0v, its just how you interpret some range CLOSE to 0v that is not exactly equal. If the hardware is *present*, and not physically removed from the circuit ENTIRELY, it will continue to consume some amount of power.

[droidhacker]

Thank you i thought it was only me.
There are so many colors so why only use shades of the same colors.
Why not use distinct colors instead?
Example: brown, black, red, orange, violet, blue, green, yellow, turquoise, pink.

I think he's already using turquoise, blue, and green.....

Edit: also red, brown, and orange...