-
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.
Penryn CPU Benchies!!!!!
- Thread starter stickroad
- Start date
Caporegime
- Joined
- 8 Jul 2003
- Posts
- 30,063
- Location
- In a house
AMD wouldn't like that Warning FUD , could be looking for a buyer soon 
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
well as was known sse3 was improved, sse4 was added, and more cache and probably a couple tweaks for latency give most of the boosts.
there is NOT a 10-30% boost in gaming performance. there is a 10-30% boost in situations at low res with settings you will NEVER< EVER use.
frankly, penryn, barcelona and nehalem aren't going to a huge amount for our gaming fps, nor do they need to. we're gpu limited, more and more cpu will simply not help there.
sse4, very VERY simply put, allows instructions flagged as sse4 to basically do 4 instructions per clock compared to 2 instructions per clock that sse3 could do. which is why its literally twice as good at it. due to the through put of data when doing an encode, more cache, better prediction both contribute the extra 15%(added to probably also some data not getting held up by the 2 clock setup with sse3.).
now as i've said all along, the one big HUGE advantage penryn COULD have had over barcelona, as there was absolutely no clear data, was if penryn could use sse4 and barca couldn't.
theres still nothing concrete, or official yet, but a few of the guys who have stated their compilers now support the barcelona also said the compilers support SSE4, which would heavily hint towards barca having sse4 instructions. in which case the boost from sse4 which artificially makes the penryn seem better than a conroe, would be extremely similar on the barca so no advantage for intel there.
the other major issue is despite the "apparent" idea that lower process instantly means much higher speeds, mobo makers seem to be getting 3.33Ghz extreme editions down to 2.33Ghz lowest clock quad cores. which isn't exactly a boost in clock speed.
its fairly up in the air which cpu will be faster after overclocking, and which will be best bang for buck(as intel could just undercut amd if Intel are slower, amd can really not do the same to intel if intel are faster). but this is utterly gaurenteed, neither new cpu will offer anything more than a 1-3% boost(maybe 5-10% in extreme and rare situations) in gaming while at a resolution that your graphics card handles fine. IE a 8800 at 1600x1200 at high detail and aa/af, which its more than capable of, still giving 60-100fps, rather than at 1024x768 with 250-400fps.
there is NOT a 10-30% boost in gaming performance. there is a 10-30% boost in situations at low res with settings you will NEVER< EVER use.
frankly, penryn, barcelona and nehalem aren't going to a huge amount for our gaming fps, nor do they need to. we're gpu limited, more and more cpu will simply not help there.
sse4, very VERY simply put, allows instructions flagged as sse4 to basically do 4 instructions per clock compared to 2 instructions per clock that sse3 could do. which is why its literally twice as good at it. due to the through put of data when doing an encode, more cache, better prediction both contribute the extra 15%(added to probably also some data not getting held up by the 2 clock setup with sse3.).
now as i've said all along, the one big HUGE advantage penryn COULD have had over barcelona, as there was absolutely no clear data, was if penryn could use sse4 and barca couldn't.
theres still nothing concrete, or official yet, but a few of the guys who have stated their compilers now support the barcelona also said the compilers support SSE4, which would heavily hint towards barca having sse4 instructions. in which case the boost from sse4 which artificially makes the penryn seem better than a conroe, would be extremely similar on the barca so no advantage for intel there.
the other major issue is despite the "apparent" idea that lower process instantly means much higher speeds, mobo makers seem to be getting 3.33Ghz extreme editions down to 2.33Ghz lowest clock quad cores. which isn't exactly a boost in clock speed.
its fairly up in the air which cpu will be faster after overclocking, and which will be best bang for buck(as intel could just undercut amd if Intel are slower, amd can really not do the same to intel if intel are faster). but this is utterly gaurenteed, neither new cpu will offer anything more than a 1-3% boost(maybe 5-10% in extreme and rare situations) in gaming while at a resolution that your graphics card handles fine. IE a 8800 at 1600x1200 at high detail and aa/af, which its more than capable of, still giving 60-100fps, rather than at 1024x768 with 250-400fps.
Soldato
- Joined
- 26 Aug 2004
- Posts
- 5,154
- Location
- South Wales
if the improvement isnt that much then why would they bother to release a new chip?drunkenmaster said:
Perfect_Chaos said:
For Intel to stay on top, and head of AMD. They feel that those tweaks are enough.
AMD k10 will be a much bigger imporvement. C2D was around 15% faster than K8. So AMD will have to surrpass C2D with K10 and be close to penryn. Also AMD need a good chip to balance the books.
Of course CPU-limited games like flight simulators (e.g., M$ Flight Simulator, Forgotten Battles), may benefit greatly. The SP1 for FSX already allows full multi-core support, which has given a great increase in FPS, so expect FSX at least to benefit from these new CPUs. Also, Oleg Maddox has in the past had close contacts with Intel during development of the IL2 series (e.g., he implemented code for smoke effects using pre-release Intel CPU features). Thus, there is hope that the up-comming Battle of Britain: Spirit of War game will take advantage of the new CPUs.