Nice, I don't think I've seen a CL2 PC133 memory module myself 😀

kixs wrote:

Finally got my GA486AM/S working with AMD 5x86-P75 in WB mode. […]
Show full quote

Finally got my GA486AM/S working with AMD 5x86-P75 in WB mode.

Config:
GA486AM/S v2.21, 256KB 15ns cache
AMD 5x86-P75 133MHz
2x4MB FPM RAM
Tseng ET6000 2.25MB PCI
BIOS set to fastest settings

Upgraded cache to 512KB (AMD@160MHz-WB):

3DBench...98.8
PCPlayer...25.6
Doom...1232
Quake...15.5

If we talking about overclocked 486, I've get faster possible results on 5x86-133mzh on 50 mhz bus:
33x4 (130mhz): 81,3 / 20,4 / 1597 / 12,6
50x3 (150mhz): 103,3 / 25,2 / 1258 / 15,1

PS but you need fast pci video card, that can run on 50 mhz pci. ET6000 for example.

philscomputerlab wrote:

Nice, I don't think I've seen a CL2 PC133 memory module myself 😀

It's laptop memory, so I'm a little dubious, but it does report as being CL2 at PC133, and CL3 at PC142 IIRC.

Phil, could you sort the table?

RacoonRider wrote:

Phil, could you sort the table?

Done!

philscomputerlab wrote:

Done!

And how about overclocked table? )

easy_john wrote:

philscomputerlab wrote:

Done!

And how about overclocked table? )

Sure

The validation for the 3DBench table is broken... no good supporting only three digits and getting people to input "Too fast" instead on the 3DBENCH field if it won't accept the "too fast" bit (also, there's a typo in the validation comment)! Just ran tests with the Dell Latitude C600 1GHz PIII and the Dell Latitude CP M233ST under Windows 98... the C600 seems to be bizarrely good at 3DBench when VGABOOST is used. I know it was a four-figure number for the 1GHz speed based on the results of the 700 MHz speed.

Thanks I'll sort it out.

Also I think this project has run long enough. We have over 500 entries, much more than I envisioned and lots of data for users to compare. So at the end of the month this project won't be taking any more entries but stay online for people to access and download.

I'm eager to start something new 😀

philscomputerlab wrote:

Thanks I'll sort it out.

Also I think this project has run long enough. We have over 500 entries, much more than I envisioned and lots of data for users to compare. So at the end of the month this project won't be taking any more entries but stay online for people to access and download.

I'm eager to start something new 😀

What's the point of killing it? Let the people add 2-3 entries a month, it's not like you have to do much.

Keep it alive 😉

Keep it alive, please.
Still have few old platform to bench and compare.

Keep it open for new entries please! 😁

philscomputerlab wrote:

Nice, I don't think I've seen a CL2 PC133 memory module myself 😀

I've got pair of them in my hands KVR133X64C2/256 they are for CUSL2_C
Also keep ita alive! 😊

Added 8 new entries!
Will add many more tomorrow!

Cannot enter "too fast" in 3dbench column!

Why is this not stickied?

Thraka wrote:

Why is this not stickied?

Dunno. We can do it if you want.

That said, if we stickied everything cool and interesting in Marvin, the first page of posts would be nothing but stickies 😀

You could make a single sticky topic that points to all the database topics. 😀 That would cut out 2 or 3 of your current stickies.

CPU frequencies in "emulators" section may be higher for CPUs having "boost" like i7 3770. As DOSBox uses only 1 core, then 3770 should work at 3.9 GHz, not 3.4. Real working frequency during test may be checked by cpu-z (as cpu-z slowes down CPU, so fps should be measured without it in memory and frequency checked in the next run).

CPUs in "stock" section should be only at native frequencies, without reduced ones. For example at line 405, I see Pentium III Coppermine E 1000 with 100x7.5. If it's not a mistyping.