coolpurplefan
Gawd
- Joined
- Feb 19, 2006
- Messages
- 547
I was wondering if the new Intel cores suffer from any weird effects like the Benny Hill gaming effect or whatever.
Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
I have never experienced the Benny Hill effect on my X2, so I would guess that it may be related to the games you play as well as your setup w.r.t. EIST/ CnQ.coolpurplefan said:I was wondering if the new Intel cores suffer from any weird effects like the Benny Hill gaming effect or whatever.
drizzt81 said:I have never experienced the Benny Hill effect on my X2, so I would guess that it may be related to the games you play as well as your setup w.r.t. EIST/ CnQ.
Wow, I have never heard of those fixes before. Would you mind giving me a link to them?coolpurplefan said:Nah, just a comment from things I read. I didn't like reading the X2 fixes because some eliminated multit-threading. I didn't make sense to me (like to shut off one core).
Fix #1 is aimed at problems that stem from CnQ slowing a single core down. Fix #2 is aimed at problems caused by CnQ slowing an inactive core down. Fix #3 is similar to #1 and #2 and only happen when "demand-based switching" is enabled. Fix #4 sets the processor affinity of a process to a single Core and Fix #5 talks about video drivers, which are very, very unlikely going to "eliminate multit-threading" on your PCcoolpurplefan said:really^ there in the AMD forum here in the X2 fix guide.
There are 4 solutions and all explained.
Windows XP processor power management implements DBS by using the adaptive processor throttling policy. [...] When single-threaded workloads run on multiprocessor systems that include dual-core configurations, the workloads may migrate across available CPU cores. This behavior is a natural artifact of how Windows schedules work across available CPU resources. However, on systems that have processor performance states that run with the adaptive processor throttling policy, this thread migration may cause the Windows kernel power manager to incorrectly calculate the optimal target performance state for the processor. This behavior occurs because an individual processor core, logical or physical, may appear to be less busy than the whole processor package actually is. On performance benchmarks that use single-threaded workloads, you may see this artifact in decreased performance results or in a high degree of variance between successive runs of identical benchmark tests.
This hotfix includes changes to the kernel power manager to track CPU use across the processor package. These changes enable visibility into the true activity level of a CPU complex and therefore help correctly calculate an increased target performance state.
Poncho said:It's funny... I've never heard of any "fixes" for Intel dual core... but have for AMD's dual core. Wonder what that means?
drizzt81 said:
- Intel processors do not report ACPI level correctly
- Intel processors do not implement ACPI states correctlyWanna back these 2 points up with some hard data?
- Intel people turn off EIST on desktop machines I don't
- Intel users are more likely to run F@H on all cores, which also eliminates the problem, since all CPUs are always at full load I don't
- ...
I do not recall ever having to resort to ANY multicore fix on my AMD based dual core PCs.
drizzt81 said:Fix #1 is aimed at problems that stem from CnQ slowing a single core down. Fix #2 is aimed at problems caused by CnQ slowing an inactive core down. Fix #3 is similar to #1 and #2 and only happen when "demand-based switching" is enabled. Fix #4 sets the processor affinity of a process to a single Core and Fix #5 talks about video drivers, which are very, very unlikely going to "eliminate multit-threading" on your PC
I am still having a hard time following? Are you talking about manually setting processor affinity?
After reading the article and the MS hotfix, it is pretty clear that the issue stems from the fact that
Given that this apparently is an OS process scheduler problem, why do you think that Intel or AMD processors would be any better at handling it?
I do not recall ever having to resort to ANY multicore fix on my AMD based dual core PCs.
- Intel processors do not report ACPI level correctly
- Intel processors do not implement ACPI states correctly
- Intel people turn off EIST on desktop machines
- Intel users are more likely to run F@H on all cores, which also eliminates the problem, since all CPUs are always at full load
- ...
I don't think that any of the issues are true. I expect the whole "dual core" problem to be a software issue. However, it appears that Poncho disagrees with my analysis. If Intel chips are indeed immune to the "dual core issue" that has been "plaguing" us poor AMD f4nbois ever since the X2 was introduced and Microsoft's hotfix documentation is indeed correct, I have a hard time imagening any other workaround that Intel may be using. I am open to be educated on the difference between AMD and Intel's ACPI implementation and how this affects their susceptibility to the "dual core issues".Dan_D said:Considering Intel helps to establish and set these types of standards, I'd like to see some data proving Intel CPU's have the issues listed above.
Summoner said:For what its worth Everquest doesnt like my core 2 duo in my laptop, i had to set it to use just 1 core and its fine then. SOE has said they dont support dual core procs though.
coolpurplefan said:thanks for posting that. I guess that means I will keep my keep a single core machine just in case (with Win XP Pro SP2).
Later on, I'll build a machine with Vista and a quad core.
EVIL-SCOTSMAN said:Well peeps, I got a strange one, amd 4400 x2 was a pain in the ass to me and most games needed some sort of fix, admittedly this was like 1 week after they were launched and better drivers have since appeared, but my 4800 x2 has never had any problem with games like my 4400 had, both cpu's used in the exact same rig and setup, so one was a pain in the ass while the other isnt, and it wasnt a faulty cpu, just that I got the 44 when x2 first came out and I got the 4800 6 months later, maybe the problem with the 44 could be put down to immature amd cpu drivers for the x2 range at the time, i dunno, but i had a nightmare of a time with the 4400....
I think most if not all dualcore problems for games have been ironed out or can be fixed by setting affinity, things have come along way since i first got my 4400 back in the middle of last year, games seem less likely to have problems with dualcore cpu's now than what i experienced back then....
Dan_D said:I doubt there is a problem with either CPU. Considering some of Microsoft's operating systems don't seem to have this issue and others do, tends to lead me to that conclusion.
drizzt81 said:
- Intel processors do not report ACPI level correctly