Epox EP MVP3G2
K6-III+500 @ 600MHz (6x100)
Geforce2Ti
Windows ME
512Mb Hynix CL2 ram @ 2-2-2-5
optimised with WPCredit

Superpi 1M in 4m 52.073s

Stedman

PII X4 955 4.2 GHz, NB 2.6GHz; GA-970A-UD3; 8GB DDR3-1333; XP SP3 32-bit - 16.344s
i3-2330m 2.2 GHz; 6GB DDR3-1333; 7 SP1 64-bit - 17.582s
PIII 900; CUBX-E; 512MB PC100 SDRAM; 98 SE - 2m 57.050s

After much frustration with being unable to get my secondary rig stable at 3.6GHz despite lots of voltage, I discovered that FSB holes are a thing. Bam, 3.8GHz 😁

Prometheus - 12.438s - 3.8GHz Core 2 Duo E8400 Wolfdale, MSI P7N SLI Platinum, 2x2GB Corsair XMS2 PC2-6400, Windows 7 x64 Ultimate SP1

CPU = Pentium 166 MMX @2.5x95
MB = m598LMR (512k Cache)
Mem = 64MB PC100 @ 95 CL3
Windows XP SP3

CPU = k6 II 533@5.5x95
MB = m598LMR (512k Cache)
Mem = 64MB PC100 @ 95 CL3
Windows XP SP3

Updated! 😁

Sub 10s
Core i7 3820 @ 4.3ghz, DDR3 2133mhz - 8.752s
Core i7 3930K @ 4.3GHz DDR3 1600 - 8.799s
-
Sub 15s
Core i5 3470S @ stock speed - 10.312s
Core i7-2600 @ 3.4ghz - 10.867s
core i7 950 @ 3.83GHz: 10.873s
Core i7 3610QM @ 3.3ghz: 11.482s
Core 2 Duo E8400 Wolfdale @ 3.8GHz , MSI P7N SLI Platinum, 2x2GB Corsair XMS2 PC2-6400, Windows 7 x64 Ultimate SP1 - 12.438s
Core2Duo E8500 @ 3.8GHZ (400FSB) - 12.476s
C2D E8500 @3,8 GHz, DDR2-800 4-4-4-12 2T- 12.558s
Core i5 760 - 12.667s.
Core i7-2720QM 2.2Ghz - 12.854s
Core2Duo E8400 @ 3.6GHz (400 x 9) DDR2-800 4-4-4-12 - 12.9s

Sub 30s
e7400 @ 3.33g(333*10), p965,ddr2-667 4-4-4-12 15.6s
Q6600 @ 3.2ghz - 15.953s
X4 960T 4.4GHz - 16.286s
PII X4 955 4.2 GHz, NB 2.6GHz; GA-970A-UD3; 8GB DDR3-1333; XP SP3 32-bit - 16.344s
Intel Pentium Dual Core E5200 @ 3333MHz (266x12.5)- 16.7356s
Phenom II 960T X6 @ 4.1GHz - 16.973s
Pentium E5200 @ 3.33GHz (266 x 12.5) DDR2-533 3-3-3-9 - 17.3s
i3-2330m 2.2 GHz; 6GB DDR3-1333; 7 SP1 64-bit - 17.582s
3.5GHz Phenom II X4 - 19.703s
Phenom II X4 965 3.4GHz - 20.686s
Phenom II X6 1055T - 21.154s
Phenom II X6 1090T -21.434s
Core i3 M330 2.13GHz - 21.513s
Phenom II 3.2Ghz - 21.699s
X4 960T 3.0GHz - 23.353s
Celeron E3400 2.60Ghz: 24.321s
e2160 @ 2.4g(266*9), p965, ddr2-533 3-3-3-9: 25s
Athlon II X3 450 - 25.672s

Sub 60s
Athlon64 3500+(venice)@2.5g(250*10), nforce3, ddr500: 34.5s
Athlon X2 4850e (2.5GHz Brisbane, 2x512k L2) - 36.219s
Pentium 4 (Cedar Mill) 3.75Gz - 36.738s
Athlon 64 3000+ (Venice E6) @ 2.5ghz - 37.172s
Prescott 3.0E @ 3.6Ghz - 38.750s
AMD Athlon 64 X2 4400+ (2.3GHz Brisbane 65nm 2x512k L2) = 39.204s
P4 651 D0 (Cedar Mill 3.4GHz) - 39.750s
P4 3.2C @ 3.6GHz (northwood) - 41.125s
Phenom II X3 E840 (1.9GHz, Caspian core, 3x512kb L2 Cache) = 41.169s
Celeron E1200 - 41.496s
Athlon 64 3200+ 2.2Ghz - 42.594s
Pentium D 930 - 42.844s
Athlon 64 3500+ 2.2Ghz - 43.906s
P4 530J, DDR2-533 4-4-4-12: 44.797s
Athlon 64 3200+ 2.2Ghz - 45.015s
Prescott 3.0E @ 3ghz - 45.094s
AMD Athlon 64 3700+ (2.2GHz 1MB L2, San Diego) = 45.312s
P4 Northwood HT 3.2Ghz - 46.734s
AMD Sempron LE-1100 - 47.750s
P4 Northwood HT 3.2GHz - 48.984s
Pentium D 940 @ 3.2ghz - 50.937s
Northwood 2.8B @ 3.15ghz - 55.57s
Pentium 4 HT Northwood 2.6GHz - 56.578s
Athlon XP 3200+ (2.2Ghz) - 57.406s
Athlon XP 2600+ (2083MHz T-bred 166MHz FSB)- 58.359s
Athlon XP 2400+ (2.0GHz T-bred) - 59.946s

Sub 3mins
Sempron 2600+ (1.83Ghz 333fsb) - 1m 05.438s
XP 2000+ (1.67Ghz Palomino 133fsb) - 1m 23.850s
p3-s 1.4g, 815e mainboard, sdram 2-2-2-5 1:36.5s
Pentium III 1.4S @ 1.6ghz: 1m 38.079s
PIII Tualatin 1400S/512/133 @ 1470MHz / 140FSB: 1m 42.387s
Pentium 3 1400S - 1m 50.764s
Athlon Thunderbird 1400C - 2m 0.766s
Pentium III 1400Mhz - 2m 6.150s
PIII Coppermine @ 1127MHz, PC133 @ 150FSB - 2m 11.278s
1.0GHz Pentium III Coppermine - 2m 21.344s
Slot 1 P3 1GHz FSB100 - 2m 28.654s
Slot 2 P3 Xeon 784MHz (7.0 x 112) - 2m 39.344s
1.1Ghz Celeron Coppermine - 2m 42.641s
2 x P3 800 - 02m 42.750s
P3 Xeon 840MHz (7.5 x 112MHz) - 2m 46.187s
Slot A Athlon 1GHz Orion - 2m 52.087s
Dual Pentium 3 Xeon 700/1Mb SL49P, ASUS XG-DLS: 2m 53.687s
PIII 900; CUBX-E; 512MB PC100 SDRAM; 98 SE - 2m 57.050s
PIII Coppermine @ 966mhz - 2m 57.883s
Slot A Athlon 950MHz Orion - 2m 58.426s
Coppermine 1000 - 2m 58.616s

Sub 5mins
Mobile Pentium 3 650 (Win2000 SP4) - 3m 02.853s
Pentium 3 650MHz - 3m 7.930s
Pentium III 933Mhz - 3m 9.189s
P3 800 Slot 1 - 03m 13.706s
P3 Xeon 750MHz (7.5 x 100MHz) - 3m 14.140s
Intel Celeron 800MHz - 3m 29.461s
Slot A Athlon 600MHz Pluto - 3m 32.616s
Dual Pentium 3 Xeon 550/1Mb SL3CE, ASUS XG-DLS - 3m 34.328s
AMD Athlon 700MHz - 3m 38.915s
Dual Pentium 3 Xeon 550/1Mb SL3CE, Intel MS440GX - 3m 48.672s
Slot 1 P3 550MHz Katmai - 3m 54.808s
Mobile Celeron 500MHz (Win2000 SP4) - 4m 6.975s
Pentium III Coppermine 600Mhz - 4m 7.143s
Intel Celeron 800 - 4m 07.752s
AMD Athlon 550MHz - 4m 13.074s
K6-III+/500 @ 600MHz (6x100) 2-2-2-5 - 4m 52.073s

Sub 10mins
Intel Celeron 466MHz - 5m 03.346s
k6-2+ @ 550(100*5.5), mvp3 board with 2mb cache, cl2-2-2-5 - 5:19
k6-2+ @ 630(105*6), ali5 board with cache disabled, cl2-2-2-5 - 5:26
k6-2+ @ 578(105*5.5), ali5 board with cache disabled, cl2-2-2-5 - 5:29
Pentium II 350MHz @ 5m 29.354s
k6-2+ @ 550, ali5 board with 512kb cache, cl2-2-2-5 - 5:41
Pentium III Katmai 450mhz- 5m 44s
k6-2+ @ 550, ali5 board with cache disabled, cl2-2-2-5 - 5:44
k6-3 @ 450, ali5 board with cache disabled, cl2-2-2-5 - 6:00
Mobile Celeron 500MHz (Coppermine, Skt 495) = 6m 00.120s
k6-3 @ 450, ali5 board with 512kb cache, sdram timing not optimized - 6:07
k6-3 @ 450, ali5 board with cache disabled, sdram timing not optimized - 6:13
k6/3 450 - 6m 18.131s
Pentium II-400 PC-100 CAS2: 6m 20.434s
Pentium II-400 PC-100 cas3 - 6m 27s
k6-2+ @ 550 (100*5.5), mvp3 board with cache disabled, cl2-2-2-5 - 6:36
Celeron 366MHz - 6m 38.563s
k6-2+ @ 450, ali5 board with cache disabled, cl2-2-2-5 - 6:50
Celeron 333MHz - 6m 57.971s
Pentium 2 266MHz - 7m 26.852s
AMD K6-2 @ 495mhz - 8m 7.024s
Celeron 266MHz - 9m 14.117
Pentium Pro 200MHz 256K Cache - 9m 41.396s
K6-2/500 @ 550MHz (half the ram not cached) - 9m 43.870s
K6-2/533 @ 522,5MHz (5.5x95) cl3 XP SP3 - 9m 51.100s

Sub 20mins
Pentium MMX 166 @ 237,5MHz (2.5x95MHz) cl3 XP SP3 - 10m 1.835s
Cyrix MII 300Mhz - 11m 23.560s
Pentium 233MMX @262Mhz (75x3.5): 12m 5.075s
Mobile Pentium MMX 233MHz (Tillamook on MMC-1) - 12m 24.011s
Pentium Pro 200Mhz - 12m 29.815s
AMD K6-233 - 12m 42.204s
Pentium 133mhz - 16m 23.153s
K6-2 475Mhz - 16m 33.135s
k6-2+ @ 133(66*2), ali5 board with 512kb cache, cl2-2-2-5 - 17:30

Sub 30mins
cyrix 6x86 @ 133(66*2), mvp3 board with 2mb cache, cl2-2-2-5 - 20m 55s
IDT Winchip C6 200: 23m 32.450s
Mobile Pentium 120MHz, NT 3.51 SP5- 25m 57.289s
Mobile Pentium 120MHz, Windows 3.11 - 27m 30s

Sub 60mins
Mobile Pentium 120MHz, Windows 98- 30m 23.536s
Pentium 66 - 30min 36sec 231ms
cyrix 6x86 @ 66(66*1), mvp3 board with cache disabled, cl3-3-3-6 - 41m 10s
Intel Pentium 75MHz - 45m 27.274s
AMD Enhanced Am486DX4 120MHz - 50m 31.356s

Sub 30 days
386SX-16 - 27days 19h 31min 43sec

Also perhaps (if theres still interest in this superpi thread, I suppose everyone lost interest as noone was updating anymore and the list became somewhat chaotic) it would be an idea to somewhat standardize the input results?
A lot of data is missing in the list which could influence the results greatly (like the chipset used, ram used and at what timings and which OS was used), so if theres any interest in a single all-including superpi list, would it be feasable to restart the list, but with more standardized results? (like in a spreadsheet or something?)

i7-4770k @ 4.3Ghz (All 4 cores) , w/ 16GB DDR3 - 2133Mhz - Windows 8.1

a classic combination:
celeron300a oc 450, bx mainboard, sdram 2-2-2-5 4:48

and a few:
pentium mmx233 oc 300(100*3), mvp3 board with 2mb cache, cl2-2-2-5 6:48
k6-2-550(100*5.5), mvp3 board with 2mb cache, cl2-2-2-5 6:50
k6-2-550(100*5.5), mvp3 board with cache disabled, cl2-2-2-5 10:15

Tetrium wrote:

Also perhaps (if theres still interest in this superpi thread, I suppose everyone lost interest as noone was updating anymore and the list became somewhat chaotic) it would be an idea to somewhat standardize the input results?
A lot of data is missing in the list which could influence the results greatly (like the chipset used, ram used and at what timings and which OS was used), so if theres any interest in a single all-including superpi list, would it be feasable to restart the list, but with more standardized results? (like in a spreadsheet or something?)

I think that'd be a good idea, make it easier to read and keep updated. Maybe like a public (shared) Google Docs spreadsheet would work. Plus it'd provide the opportunity to re-run the tests which is always fun 😁

Joined this amazing forum for retro pc's today.

SuperPi 1M time 3m 03.163s - PIII Coppermine 850 MHz (8.5 x 100) / MSI MS-6368 ver 2.0(VIA Apollo PLE 133)/ 512MB PC133 SDRAM (3-3-3-6)/ WinXPSP3

I think it would be also good to differentiate between Windows 9x and windows "NT" results,
also always specify motherboard (and cache), fsb, memory settings (as some are doing, but not all), that way the times would make more sense, but yes... it's hard to organize and interest in this is looking low at the moment,

my latest run with the PII 400 with c2 ram (a few other lower latency settings for memory) was completed in 5:47 (still with win98) compared to my previous run with c3 and 6:27 for example.

SPBHM wrote:

I think it would be also good to differentiate between Windows 9x and windows "NT" results,
also always specify motherboard (and cache), fsb, memory settings (as some are doing, but not all), that way the times would make more sense, but yes... it's hard to organize and interest in this is looking low at the moment,

my latest run with the PII 400 with c2 ram (a few other lower latency settings for memory) was completed in 5:47 (still with win98) compared to my previous run with c3 and 6:27 for example.

So how would we go about?
So how about this (I'm fabricating this result, just as an example):
Intel Pentium MMX 233 (insert CPU core here), 233MHz (3.5x66) (insert actual overclocked speed here if applicable, otherwise insert "stock"), Windows 98SE, Superpi (insert version), Mainboard (including revision, chipset, amount of cache etc), RAM (including latencies etc), Superpi result in days, minutes, seconds (including milliseconds if applicable).
So we end up with something like this:

* Intel Pentium MMX 233 (P55C), 233MHz (3.5x66), 250MHz (2.5x100), Windows ME + unofficial SP, Superpi modded2, ASUS P5A rev 1.04 (ALi M1541 Aladdin V + 512KB), SDRAM PC-100 2/2/2/5 128MB, 13m 37.123s

Above results are fictional, fabricated and fake 😜 It's merely an example.
Btw the Superpi modded2 is the superpi that can be downloaded in the first post in this thread. It displays milliseconds in Windows 9x but has a 1st run crash bug when being run when theres no prior results. The one I uploaded (is the same one in the very first post) has 1 dummy result already so it won't cause a crash anymore and is nice because it's the only Superpi I know of that will display milliseconds and works on Windows 9x

The thing also is to make the results somewhat nice to read while not leaving out any info that might be relevant.
I have some limited knowledge about spreadsheets, but I'm definitely not a master at this but I'm willing to learn if this is a good idea and I have the support of the people (god I guess I played a bit too much RTW lately 🤣!)

Also it "might" be a good idea to start a new thread if we decide to make a spreadsheet version of the all amazing super duper Superpi results thread.
I was thinking this: Everyone who wants to (and doesn't fear he's gonna mess up the spreadsheet 😜 ) may edit the spreadsheet.
However, if you decide to be the one updating the spreadsheet, please post so before actually updating the spreadsheet. This is to prevent multiple users from trying to edit the same document at the same time.
So my suggestion is to do it like this:
*Make a post (can be with your own new Superpi result) and state that you are going to update the spreadsheet
*Edit the spreadsheet adding all the results not yet added since the last time the someone declared he was going to edit the spreadsheet the previous time
*After saving the spreadsheet, edit your post where you stated you were going to edit the spreadsheet and mention that the spreadsheet has been edited and all results up to your own post have been updated.

Good idea?

Please add if I missed anything 😀

This must be pretty fast for an old socket 775 rig 😀
I had to up the CPU clock 160 MHz over this systems normal 24/7 Prime stable OC settings to get under 10s.
The CPU is stable at 4660 but it needs a little bit extra VTT to be able to Prime.
I am very careful when it comes to VTT and high end Wolfdales/Yorkfields, I like to leave it at default.

All settings in the BIOS are set to optimized defaults except the FSB, memory timings and divider, memory voltage (2.1V) and CPU voltage (1.45V).
SpeedStep is disabled. The rest of the voltage settings are set to the "normal" setting not the "auto" setting.

That is a freakin fast c2d. You're at the same level as an X79!!! 😲

Also, good idea on the spreadsheet, but I can see it's not been taken up?

sgt76 wrote:

That is a freakin fast c2d. You're at the same level as an X79!!! 😲

Also, good idea on the spreadsheet, but I can see it's not been taken up?

I might do it...if I find the time. The thing is, I have no idea how to do this, so I'd need to start doing/learning it from scratch.

Core i7 4930K @ stock (3.4GHz) - DDR3 1866MHz - 9.422s

...not overclocked unlike the other sub-10 results up there (4.3GHZ!! 😳)

Tetrium wrote:

sgt76 wrote:

That is a freakin fast c2d. You're at the same level as an X79!!! 😲

Also, good idea on the spreadsheet, but I can see it's not been taken up?

I might do it...if I find the time. The thing is, I have no idea how to do this, so I'd need to start doing/learning it from scratch.

it would be a challenge given the effort required. maybe starting with some guidelines as you ve suggested plus starting a new list may be a logical step fwd

Celeron 1GHz @ 1235MHz 2,05V 123.5MHz FSB 02m 50.710s
PIII 800B @ 988MHz 1,90V 123.5MHz FSB 02m 50.817s
PIII 866EB @ 863MHz 1,70V 132.7MHz FSB 02m 58.915s
PIII 866EB @ 863MHz 1,70V 132.7MHz FSB 02m 44.863s (installed Intel Chipset driver and enabled DMA on harddrive)
PIII 866EB @ 863MHz 1,70V 132.7MHz FSB 02m 41.387s (same as above, but whith RAM running 2-2-2 timings)
PIII 866EB @ 863MHz 1,70V 132.7MHz FSB 02m 41.018s (same as above, but using ramdisk)

Intel SE440BX-2
Asus slotket
256MB SDRAM @ 3-2-2 timings
Windows 98SE
Overclocked with CPUFSB 2.2.6 (Intel SE440BX-2 have no settings for FSB and 133MHz FSB CPU's are "downgraded" to 100MHz)

Edit: Added another result and yet another one.

Core 2 Duo E6700 @ 3 GHz:
1M = 21 seconds

486DX/33:
128K = 16m 6s

Shaved off another 6 seconds on my 866EB with some additional memory tuning. Added a better heatsink to the 800B som it could finish Super PI at 1062MHz. 😀

PIII 866EB @ 863MHz 1,70V 132.7MHz FSB 02m 34.963s
PIII 800B @ 1062MHz 1,90V 132.7MHz FSB 02m 19.955s

Intel SE440BX-2 with P13 BIOS
Asus slotket
256MB SDRAM @ 2-2-2 timings
Windows 98SE
Overclocked with "CPUFSB 2.2.6" (Intel SE440BX-2 have no settings for FSB and 133MHz FSB CPU's are "downgraded" to 100MHz)
Tuned memory with "WBXTUNE 2.31" (DRAM Idle timer: Infinite | Idle/Pipeline DRAM Leadoff Timing: 0 wait | Host Bus Fast Data Ready: Enable | DRAM Refresh Rate: 249.6us)
Running "Super PI" from ramdisk.

Tried same settings with Windows 2000 SP4 instead, without ramdisk.
PIII 800B @ 1062MHz 1,90V 132.7MHz FSB 02m 06.552s (FAT32)
PIII 800B @ 1062MHz 1,90V 132.7MHz FSB 02m 02.126s (NTFS)

Edit: Added the 800B run. Added Windows 2000 run.