Dell GX270, socket 478 p4 2.6. It has an option for "compatible" cpu speed that slows it to a 486-66. However a timedemo in doom is all wrong when comparing to the benchmarks in the "benchmark with caches disabled" thread. Similar to what happens when I use slowdown programs, the demo runs in slow motion (not jerky) and not at all "as fast as it can go" like a regular timedemo. Interesting nonetheless though.
Does this DELL have an option to disable L1 Cache in the BIOS? I'm wondering if what you are seeing is the equivalent to this?
Disabling L1 Cache on Pentium 4 class CPUs results in around a fast 486 CPU. I've tried this with Prescotts on i865/ i875 chipsets.
Do you have any benchmarks to post for this DELL? E.g. Speedsys.
wrote:Does this DELL have an option to disable L1 Cache in the BIOS? I'm wondering if what you are seeing is the equivalent to this?
Inside the BIOS Setup: CPU Information -> CPU Speed: Normal/Compatible, Just checked on my Dell Dimension 4300S Desktop, Should pretty much be the same on any Dell Desktop from 2001-2004.
wrote:Do you have any benchmarks to post for this DELL? E.g. Speedsys.
Even though I'm not the OP, I can bring up the TOPBENCH scores I've received in both CPU modes on my 4300S for comparison purposes.
TOPBENCH Score in Normal Mode:311
TOPBENCH Score in Compatible Mode:97 <- Closely matches to a 386DX-40
Sorry it's not in picture form ATM, I'm in a bit of a rush since this is my first time being here and I just wanted to answer some questions.
wrote:Does this DELL have an option to disable L1 Cache in the BIOS? I'm wondering if what you are seeing is the equivalent to this?
Disabling L1 Cache on Pentium 4 class CPUs results in around a fast 486 CPU. I've tried this with Prescotts on i865/ i875 chipsets.
Do you have any benchmarks to post for this DELL? E.g. Speedsys.
Topbench crashes, speedsys hangs on "get dmi info".
I managed to run topbench on it somehow one time, and got the 486/66 figure. The benchmark with a red screen rates the cpu at 105 and fpu at 565... is there any reason why topbench and speedsys would crash when being run from cd?
@KCompRoom2000:
Thanks for the benches. What's the spec of the actual CPU being used, for reference? Have you tried Setmul to see if you can disable L1 cache, and compare benchmarks to see if this is equivalent to the BIOS setting?
@bluejeans:
Thanks, too, for the benches. What's your CPU spec? Also, as above, have you tried Setmul?
BTW, regarding your failed benches, usually some BIOS setting is causing this. Try disabling anything to do with caching Video BIOS or System BIOS options, for example.
wrote:@KCompRoom2000: Thanks for the benches. What's the spec of the actual CPU being used, for reference? Have you tried Setmul to […]
@KCompRoom2000:
Thanks for the benches. What's the spec of the actual CPU being used, for reference? Have you tried Setmul to see if you can disable L1 cache, and compare benchmarks to see if this is equivalent to the BIOS setting?@bluejeans:
Thanks, too, for the benches. What's your CPU spec? Also, as above, have you tried Setmul?BTW, regarding your failed benches, usually some BIOS setting is causing this. Try disabling anything to do with caching Video BIOS or System BIOS options, for example.
P4 2.6ghz northwood. Dell optiplex gx270. I thought setmul was only for certain boards and k6-3 cpu's? Also, I don't think video bios or system bios caching is even anything you can turn off on this proprietary system. Btw, whenever I've tried this, or other slowdown programs, doom timedemos run really slow, a lot slower than normal gameplay. Why might that be?
wrote:@KCompRoom2000:
Thanks for the benches. What's the spec of the actual CPU being used, for reference? Have you tried Setmul to see if you can disable L1 cache, and compare benchmarks to see if this is equivalent to the BIOS setting?
The Dell Dimension 4300S that I've posted benchmarks of has a 1.6 GHz Intel Pentium 4 Willamette CPU.
wrote:is there any reason why topbench and speedsys would crash when being run from cd?
I'm guessing the reason why TOPBENCH doesn't work from your CD is because when ran in full mode, It tries to write the list of profiled systems to the disk which is impossible on a CD-R since that's a WORM (Write once, read many) media; have you tried running it with the -i switch? that should print the system's score into the DOS prompt so there shouldn't be a problem with that method, Worst case scenario you can try running it from a floppy disk or a USB flash drive.
wrote:The Dell Dimension 4300S that I've posted benchmarks of has a 1.6 GHz Intel Pentium 4 Willamette CPU. […]
wrote:@KCompRoom2000:
Thanks for the benches. What's the spec of the actual CPU being used, for reference? Have you tried Setmul to see if you can disable L1 cache, and compare benchmarks to see if this is equivalent to the BIOS setting?The Dell Dimension 4300S that I've posted benchmarks of has a 1.6 GHz Intel Pentium 4 Willamette CPU.
wrote:is there any reason why topbench and speedsys would crash when being run from cd?
I'm guessing the reason why TOPBENCH doesn't work from your CD is because when ran in full mode, It tries to write the list of profiled systems to the disk which is impossible on a CD-R since that's a WORM (Write once, read many) media; have you tried running it with the -i switch? that should print the system's score into the DOS prompt so there shouldn't be a problem with that method, Worst case scenario you can try running it from a floppy disk or a USB flash drive.
I somehow got it to run when I first discovered the feature, and it did detect as a 486/66.
