VOGONS


The World's Fastest 486

Topic actions

Reply 300 of 320, by kool kitty89

User metadata
Rank Member
Rank
Member

Hmm looking at the list here:

https://thandor.net/list_object/family/Am486

The 25253 parts number was used for 5V rated DX4-100 parts as well as the 3V DX2-80 and DX2-66, so it should tolerate 5V OK but may not have liked the particular boards and/or didn't like running at 3x.

I'll keep that in mind for the Acorp board since it's stuck at 2x for now along with only 3.46 and 5.1V settings.

Reply 301 of 320, by The Serpent Rider

User metadata
Rank l33t
Rank
l33t

The 25253 parts number was used for 5V rated DX4-100 parts

No such thing.

Get up, come on get down with the sickness
Open up your hate, and let it flow into me

Reply 302 of 320, by Thandor

User metadata
Rank Member
Rank
Member
The Serpent Rider wrote on 2020-05-28, 10:35:

The 25253 parts number was used for 5V rated DX4-100 parts

No such thing.

kool kitty89 saw on my website that the DX4/100 (25253) runs at 5V but I've mistakenly mentioned 5V for this particular CPU. The image of my Am486 DX4/100 clearly shows 3V. I've corrected the information on my site.

So the 25253 runs at 3V 😀.

thandor.net - hardware
And the rest of us would be carousing the aisles, stuffing baloney.

Reply 303 of 320, by kool kitty89

User metadata
Rank Member
Rank
Member

I assume all the Intel DX4 Overdrive CPUs at 5V were using voltage regulating interposers.

I also wasn't able to test any of those AMD DX2s or DX4s in the Acorp board as its socket seems to have gotten a bit flakier. Either the ZIF locking mechanism was already slightly damaged before I got it and causing mis-allignment with the underlying contacts or the contacts have gotten bent since then and are making poorer contact.

It may be related to using CPUs with formerly bent pins that are still slightly off, but at some point the socket also started bending CPU pins at odd angles where they were straight beforehand. I may have to try to disassemble the socket at some point and adjust/repair the contacts. I'd like to avoid trying to desolder replace the socket if possible.

Or it's both issues combined: mis-alligned socker cover plate causing the internal contacts to be partially exposed when inserting and thus causing things to get bent. There's more friction than there should be while inserting CPUs, so that may be it. That or there's just a variety of crappier or poorly designed ZIF sockets from the mid 1990s that don't behave like typical Socket 5 or 7 ones.

The Soyo SiS 471 board has a finicky socket, too, and like the Acorp one also has cylindrical recesses in all the holes, so it's easy to mis-allign them and end up bending the pins slightly. Better sockets have conical recesses that help guide the pins in. (the turned brass style non-ZIF sockets also tend to have tapered, conical recesses around the pin holes that help guide them in)
Those blue Intel Overdrive style sockets seem to be like that and some white/beige colored Socket 3 ZIF ones that behave pretty much the same as typical Socket 7 fare. (that other SiS 496 board does have such a socket and I'm more confident in it)

I may have also been correct about the QFP mounted Cx5x86 having larger/thicker pins than the others and may have contributed to making other CPUs harder to make contact in the Acorp board. (and the stiffer pins are less prone to get bent by the socket's weirdness)

After frustrating fiddling with several AMD DX2-80s, DX4s, DX5s and several of the Cyrix DX2s and DX4s I'd been using before not POSTing and ending up with pins bent on removal, I went back to the Cyrix 5x86 and it's working fine as before, both at 120 MHz and 3.46V and at 133 MHz and 5V, but I haven't put much more time on it in the latter configuration.

The trouble really seems completely mechanical in nature with the socket and not any electrical stability issue or damage. Though it's possible the higher ambient temps caused the socket to expand/warp slightly and behave worse than it did in my night-time and early morning testing. (the problems arose in the afternoon after I'd been successfully playing with the ST DX4 at 2x66 and 5V for a while and had Jazz Jackrabbit's attract mode idling in the background ... and I'm still much more confortable running that chip at 5V than the 5x86 and certainly the AMD .35 micron parts)

I also managed to partially corrupt my DOS 5 install when doing some of my benchmarking at more experimental settings, but didn't seem to loose any data (but some files in the DOS folder got corrupted, some commands and the DOS shell stopped working). Then during a separate batch of tests in the Symphony board again, I managed to corrupt/delete my Topbench results by attempting to save and exit and having the system crash/freeze and when reverting to a known-stable configuration, TOPBENCH showed a totally empty results archive. (and I stupidly haven't been backing them up regularly, though I had a backup from about a month ago I just used to repair DOS as well as replace the empty topbench archive)

On the plus side, I have screenshot photos of most of the interesting stuff, but the archive compilation itself was nice.

It also turns out that crash was caused by cache instability ... I'd been trying the 43.75 MHz FSB configuration again, this time without cache wait states on and didn't provide extra cooling to the cache. Trying that again later, after backing up the repaired DOS install was actually successful, but still not a stable configuration for running most programs. Also it's not objectively impressive, but for a 1991, ISA only board it's quite good. OTOH with cache wait states enabled at 131 MHz the ST486DX-4 is about the same (or slower for some things) than at 4x40 MHz, and the board as a whole is more stable there. There's also no way I can see to add write cycle wait states, so DRAM stability at higher bus speeds is also difficult, and it seems to be more finicky with 4MB SIMMs than 1MB, but I've only got one or maybe 2 1MB 60 ns simms but more than a dozen 4MB ones, and sorting through the fastest 70 ns modules I had still didn't quite get there. (I'm not sure how I could get it to actually do 50 MHz FSB as it's marked for)

And while the Cyrix based DX2 and DX4s (and shot period testing of the 5x86) all seemed pretty solid performers in the Symphony board and the actual 5V rated AMD and Intel DX2s and Intel DXs all seem to go OK, it doesn't overclock the AMD and Intel CPUs as well as the Opti or SiS based boards (the AMD DX2 66s tend to run fine at 80 MHz in the other boards and Intel DX2-50 does fine at 80 MHz as well) they do run fine at stock settings or sometimes at smaller overclocks (the DX2-50 is fine at 66 MHz). However, the .5 micron AMD CPUs seem to have weird issues that may be related to running at 5V or may be BIOS or chipset related, plus the 2 DX2-80s are stuck in 3x multiplier mode which would be fine if they overclocked even a little or even went OK at 27x3, but in all cases they threw errors when loading Quake, though most other things worked OK.

OTOH that other SiS 496 based board handled the .5 micron AMD chips fine and overclocked the DX2-80s to 3x40 MHz at 3.98V and would POST at 150 MHz for the AMD DX4-100 but wouldn't boot anything. If I could find the 2x multiplier jumper I suspect it would do 2x66 MHz, though, and as it was seemed to be happy at up to 120 MHz at lower voltages, possibly down to 3.3V if kept really cool, but I did most of the testing at the 5.51 or 3.68V settings.

I also confirmed that it is indeed something related to the vintage of board/BIOS (1994) that makes it behave badly with 3x multiplier Cyrix CPUs and is presumably attempting to set-up certain things at non-default settings given those same CPUs work better in even older boards with no support for Cyrix 486DX2 let alone DX4 CPUs. All the Cyrix/IBM/ST 486 DX2 models I tried in that board worked properly, albeit limited to running at 3.98V max, which undervolted several of them. (interestingly, the 5V ST DX2-80 which will not even POST at 100 MHz, seemed to be fine with 3.98V at 80 MHz, though I've seen one or two other CPUs that are a bit like that: tolerant of under-voltage but poor overclockers)

So at least without a BIOS update, that board is better for the AMD x5 CPUs since they at least get detected as Enhanced AMD 486 DX4s and seem to work properly, though are set to write-through mode for some reason. (I don't know why since Enhanced AM486s had WB cache and should either have that enabled or have a selection option in the BIOS ... since the BIOS does properly identify that feature at POST)

I suppose it could also be something weird related to the GUI BIOS of that board, that and the very similar one of the Soyo board which won't POST any DX4 CPUs at all.

I think I also forgot to mention that the Acorp board does have an L1 cache policy select option in the BIOS, but it only appears when Cyrix CPUs are installed and disappears for others. The GUI BIOS of the Soyo SiS 471 and other 496 board have those options grayed out and unselectable with non-Cyrix CPUs but still visible. (though, again, they don't seem to work properly with Cyrix DX4 or 5x86 CPUs while DX2s are fine and I don't have any working DX or SX 1x multiplier models to test)

My take-away so far for Cyrix/IBM/ST 486s is that they at least seem to be more compatible and reliable with older boards. On top of that, they had some of the fastest, formally 5V rated/tolerant CPU models available. On top of that, they have faster external memory performance than AMD and Intel 486s (at identical settings) and seem to tolerate tighter cache and DRAM timings than AMD or Intel parts, further widening that gap.

The external I/O bandwidth seems to push the Cyrix 486s ahead in Speedsys scores and some other benchmarks (probably most productivity related ones, but also some graphics/games ones like PC Player: the 640x480 VESA mode PC Player benchmark seems to really like fast memory performance)

Ignoring the overvolted ST DX4 CPUs, the Cyrix/ST 5V 486DX2-80 should have been a really appealing upgrade option for Socket 1 boards capable of being configured at 40 MHz. That and the 5V 66 MHz rated DX2s seem to overclock fairly well along with the 3.x V ones (and I'm still not sure which the IBM BL DX2 is). Plus you still get another big speed boost if you upgraded the board later on and re-used the old CPU in a 1996 vintage PCI or VLB board.

Plus, if you had decently fast RAM there was still the SIMM extender/expander/saver style adapter route. That's what my Dad used in a couple of our 486 or Socket 5/7 era builds: actually most of those modules are in storage, still populated, I think with 70 ns FPM DRAM. (not sure what we used for video cards: as of 12 years ago, we had no VLB cards in storage, a single ATi Graphics Ultra ISA card, and the oldest PCI video card was a Diamond Stealth 2000 S3 ViRGE 325 with 1996 dated BIOS V1.01)

That Virge also does fine at 40 MHz but will not do 50 or higher without issues, mostly with VGA palette corruption, possibly specifically limited to the RAMDAC's palette RAM. 256 color VESA/SVGA modes seem to avoid this entirely ... so Tomb Raider at 640x480 works fine, as does PC Player, and same for Terminal Velocity, but anything using mode 13h or unchained VGA mode X/Y end up with corrupt colors but not framebuffer RAM errors. Presumably 320x240 using VESA graphics modes and not VGA-compatible modes would also avoid the palette errors and also use the same linear framebuffer organization as 320x200 VGA 13h and 640x480 VESA.

I'm not sure what's going on with Quake as it's missing the VESA resolution options in DOS but has most or all of them in my Windows 98 build ... that or all those are just really high-res unchained VGA framebuffer modes. (does Quake insist on having enough video RAM for 4x pages/framebuffers like Doom uses? ... in that case the limited resolution options with 512k 1MB or 2MB video cards would make more sense, but still not explain all the restrictions I see ... or maybe an issue with me running the cards driverless in DOS)

I'll have to try S3D accelerated games at 50-66 MHz PCI at some point or DirectX.

I assume as sluggish as it can be, the S3D version of Tomb Raider is faster than software rendering at 640x480 on a 486 in the 80-120 MHz range, especially if there were actual detail setting options or command line tweaks. (Tomb Raider tends to look a bit weird with texture filtering anyway ... even Tomb Raider 2 does, but perspective correction and highcolor rendering is nice ... or well, I'd probably prefer nicer shading and lighting with 320x240 24-bit truecolor mode on the ViRGE ... I do in Tomb Raider 2 except it's got those weird rendering errors around the HUD and health bar in Direct X)

Reply 304 of 320, by kool kitty89

User metadata
Rank Member
Rank
Member

I just realized that I could potentially also work around finicky sockets + bent/damaged pins by using a PGA-168 socket as an adapter.

I was already considering getting a lot of those for potential 486 socket replacements on damaged boards (albeit if replacing a Socket 3 board with that, you lose Pentium Overdrive compatibility) but this could make them even more useful.

I've actually had a good deal less trouble with the turned brass style PGA-168 sockets than some of these ZIF designs, so that's a good sign.

I do have a Biostar 486 with a turned brass style Socket 2 or 3 PGA socket too, which is nice, though it needs a new BIOS. (I've got an image for this one, actually, but need to burn an EPROM for it)

And I tried other CPUs in that Acorp board again, but even with some fiddlying with insertion angle and working the lever a bit, no luck other than the Cx5x86 in there.

Also, the 3 ceramic Cyrix 5x86-100GP chips I have turned out to all be Stepping 1 revision 3, vs the Stepping 0 Revision 5 QFP mounted one I've been using. (they're also in what looks like new old stock condition, though I forget what the seller listed them as)
I only tested them in that SiS 496 board that doesn't want to work right with 3x multi Cyrix 486s or 5x86s, so L1 cache was stuck disabled, but I could still get into DOS for the CPU ID info in Chkcpu and run various things slowly.

They all have matching parts numbering on the back, which makes sense, though looking at the other damaged QFP mounted 5x86s I got last year, there's a variety of different identification codes on them, but I'm not sure what they correspond to and most of those are non-working though probably not dead. (just pinched/shorted or lifted leads which should be able to be repaired, since they look like they were damaged in storage and/or during handling in e-waste recycler processing or some such ... though they don't look like they ever had heatsinks attached, so I get the impression they weren't installed in boards)

Hmm, plus all those QFP mounted CPUs are missing the decoupling capacitors, so either they were unfinished left-over factory stockpile or just cheaper mounts not ever destined for caps. (I'm pretty all the QFP adapter mounted 5x86s I've seen elsewhere have had the caps populated, or at least AMD x5s and AMD/Intel DX4s mounted like that)

Edit:
Of those QFP mounted 5x86-100GP CPUs, 2 are working as-is: the one I've been testing and another, which I originally POST-tested in the OPTi 495SX board. The latter doesn't want to POST in either of those SiS 496 boards, but seems to work fine in the Symphony 386/486 board, so I'm not sure what's going on there. (might just not being making good contact with either of those ZIF sockets ... and I did try the Acorp board at 5V to be sure there: thinking it might work with other QFP adapted CPUs with the same pin types, but I think it's just a matter of mechanical fit there and that other 5x86 happening to make good contact)

I have no reason to assume that other 5x86 is damaged in such a way that it prefers/needs 5V and it probably simply tolerates it like the first one I've been testing.

Both of these are Stepping 0 rev 5 CPUs according to CHKCPU.

Reply 305 of 320, by kool kitty89

User metadata
Rank Member
Rank
Member

I still need to post pics related to the last few updates, but I dug into a Biostar MB3486 with UMC UM82C481/482 chipset.

While it annoyingly seems to have no parity disable option, finding stable configurations with parity errors in mind isn't that much more difficult and I don't have that much SIMM-30 RAM that's got parity-specific issues (slow, bad, or absent parity chips or damage to bit 9 on the connector).

It seems to be the fastest of the 3 386/486 boards I've tried even being ISA only (there's an unpopulated spot VLB style slot, referred to as Local Bus MCA in the manual, but the ISA slot itself has problems there, so I'm not sure about noise/stability on that area of the board).

It's got a socketed master oscillator, too, except without the divider for common 386 and 486 clock rates. So 33 MHz 386 needs a 66 MHz oscillator, but 33 MHz 486 (or 486 bus) needs 33 MHz.

But as I just discovered, it's actually capable of running a 66 MHz 486 FSB without apparent issues (with ISA set to 22 MHz for fast video) albeit the only CPU I have capable of that is the Intel DX-50 I had running at that speed previously. (using an interposer with 2x/3x multiplier jumper I'd have more options, but the system as a whole would probably be faster at 3x44 than 2x66 MHz due to lower wait states)

One of the Cyrix DX-2/66s (one of the Stepping 3 revision 2 ones marked for lower voltage) manages to run at 2x54 MHz with good results in this board and the ISA bus actually seems to work fine at 27 MHz with wait states enabled on the CL5422. (even the Vibra 16 which usually has problems with Jazz Jackrabbit or Doom's sound doesn't have issues with that set-up)

And the most heavily used/tested ST DX4-100, while still not POSTing at 3x50 MHz actually seems to run fine at 3x48 MHz without signs of trouble or weird finicky behavior, so that must be right on the edge of what it's capable of. (that and it's probably got better stability in the L1 cache RAM+logic+IO region than some other examples, given that's often the breaking point)
Albeit the benchmarks I tested it in tended to specify slightly under 48 MHz, usually 47.9 MHz, so the actual clock speed might be just under 144 MHz.

OTOH the 66 MHz oscillator I'm using is 66.0 and not 66.67, for what that's worth. (the oscillator the board came with also appears to be 33.0 rather than 33.33)

The memory controller in this UMC chipset also seems to be quite fast, similar to the Symphony one when L2 cache is enabled, but much faster when the L2 is disabled, so I assume it's stuck in write-back mode like the Opti 495SX. (though unlike the 496SX board, there was no 2-2-2-2 cache timing option which is unfortunate given that seems to be particularly well matched to Cyrix CPUs)

That Intel DX-50 combination with this board is kind of neat in particular given it's a 1991 board with a 1991 CPU (though I'm not sure on the manufacturing date of that particular DX-50) mated to a pretty substantial 32% overclock and at stock rated voltage.

Granted, the 66 to 108 MHz overclock is a more impressive 62% jump, but that's also on a 3.6/4.0V rated CPU. (judging by the stepping and revision numbers I'm seeing, Cyrix DX2s of Stepping 3 from revisions 1 to 4 appear to have been used for both low voltage and 5V versions)

I think this got mentioned here or in another thread before (maybe the Ultimate 486 comparison thread), but I noticed the Stepping 0 Revision B (so rev 11 in decimal) benches slightly faster at the same clock speed and settings in the same board as do the Stepping 3 rev 1, 2, and 4 CPUs I have. (the difference between Speedsys scores of 31.98 and 34.04, both at 2x40 MHz with the same settings in the Soyo SiS 471 board)

I might be mixing this up with a discussion on another model revision, but I recall Feipoa mentioning something about Cyrix improving compatibility, stability, or yields at the expense of performance.

Reply 306 of 320, by kool kitty89

User metadata
Rank Member
Rank
Member

Correction to any previous reference I made to the MB3486 motherboard. I got my wires crossed: it's a US 3486 board (based on the manual naming). The UMC chipset listed is correct, but it's not a Biostar board.

I've also gotten an Intel 486SX-33 to run at 66 MHz in it with similar results to the DX-50 and am currently running a TI486DLC-40 at 50 MHz along with a IIT 87DLC-33. (I'll dig out my Cyrix 87DLC to compare later on, though I recall them being pretty close in other comparisons)

An AMD 386DX-40 I tried also ran fine at 50 MHz, while it wouldn't POST above 40 in the Symphony board, so that might be a good indication of that board's limitations being stretched at 44 MHz. (the 486DLC fared better as do most PGA-168 CPUs, so it's a bit confusing)

And unlike the Symphony board's odd behavior with the DLC (apparently no L1 cache enabled, also Speedsys freezing, but most other things working fine) the 486DLC seems to be totally normal and functional in the US3486 board, in spite of the manual not mentioning Cyrix CPUs at all.

However, the board-level cache is so fast that it totally masks the 1k DLC cache when enabled, with the on-chip cache being very slightly slower (at 50 MHz in Cachecheck giving 27 us vs 26 us for the L2) and shows up with a very slight dip for the first 1k in Speedsys. The BIOS also doesn't allow the external cache to be enabled alone (only internal, none, or both) so I can't be sure here, but I'm going to assume overall external I/O performance would be poorer with the internal cache disabled as it tends to be with most 486s.

Also, while Speedsys and a few other benchmarks only show about a 50% improvement over a similarly clocked 386DX, Landmark's score is much more DLC-biased and the 3D games I have installed seem much more DLC favorable as well. (plus Tomb Raider continues to refuse to run on any pure 386 ... so in spite of some anecdotes I've read, it seems to need something 486-specific to work: otherwise it always crashes while loading a save file or starting a new game)

Quake works on a vanilla 386 with FPU, so that comparison can be made, but Tomb Raider is at least somewhat playable on lower or minimal settings and the sound engine doesn't go all choppy as in Quake. (presumably Tomb Raider would run better than Quake does on a 386DX if it would actually work, and the 3D menu runs OK on a 386 and also isn't impacted by the presence of an FPU where the in-game engine runs extremely slowly without one ... I can only imagine it uses FPU emulation that might be oriented towards Nx586 systems as it seems pointless on a 486DLC or SX)

So Doom, X-Wing, Tie Fighter, Terminal Velocity, and Quake seem to be closer to Intel/AMD 486 performance and well above the 1.5x gain over the 386. (albeit that's not always related to average framerate output, sometimes more an issue of higher minimal framerate and less slowdown) X-Wing's detect.exe gets around a 2x gain for the DLC over the 386. (486

Oh and Doom at max details gets 5135 realticks while Quake manages to get 2.9 FPS with the DLC-50 and IIT 87DLC with the CL5422 (1MB DRAM) with WS enabled and ISA clock at 25 MHz. (or in other words, Quake is basically just a 3D demo, not a game and Doom is better played at fullscreen in low detail mode)

And interestingly, Quake only improves to 4.5 FPS at the smallest screen size, so I'll bet it's choking on the geometry-related stuff and/or hidden surface removal rather than the FPU-driven texture span rendering. This is something I've noticed 486DXs appear to struggle with as well.
(which implies to me that a special low-detail mode might have helped without needing to modify the texture span pipeline itself: like an integer-driven geometry engine and/or disabling hidden surface removal and relying on the low bandwidth and reduced overdraw penalty at low res, like using double-wide and double-tall pixels, which I think is what that Quake adaptation on the Atari Falcon uses)

I'm actually starting to wonder if Tomb Raider uses the FPU for the texture pipeline as well, while relying on integer/fixed-point math for the geometry. Use of 16-bit fixed-point geometry might explain the tendency for gaps and seams to show up so often between polygons, similar to the Playstation version. (supposedly that sort of sub-pixel accuracy artifacting can be avoided with careful attention to rounding errors, but rarely gets seen in practice)

Actually, I forget how favorable the 16-bit multiply is on the P5 Pentium vs its fast floating point multiply. I know 32-bit integer multiply is slower, but 16-bit math would be more relevant to fast 3D on the 386/486, and you get further gains from fast 16 or 32-bit adds and divides relevant to 3D matrix math. So Tomb Raider might be mutually 486 and pentium friendly.

Most of the bugs/issues that 25 MHz ISA has with Socket 1/2/3 486s in certain games all seem to be absent with the DLC, no need to add I/O recovery time wait states either. X-Wing's sound engine seems fine at fastest I/O timings, Doom's sound effects aren't distorted, and Jazz doesn't crash. I assume the DLC just isn't fast enough to run into the problems faster 486s do, but it's handy for getting good video bandwidth without VLB and without fiddling with 50 MHz VLB quirks. (and for 16-bit software, the video bus activity should be 16-bit wide as well, so the 32-bit bus width of the 486 local bus won't help)

For whatever reason, the ISA bus also seems faster than on other 486 boards I've tried, both the other Socket 1 boards and Socket 3 boards, including comparisons at the same overclocked ISA speeds. With faster (mostly 100+ MHz) 486DX2 and DX4s 3DSpace can get up to 98%, which is up there with the PCI based AMD x5 systems or PII/PIII or higher territory. (where you're running up against the 99% barrier) Where the ISA or 'slow ISA' wall for that benchmark looks like around 50-55%. (the 486DLC can manage 54% in this board)

I did try some comparisons at the more acceptable 8-10 MHz ISA clock range, mostly with 0WS enabled, and performance does drop off, but I also noticed stability is a lot harder to maintain with 0WS enabled. (ie it took more tweaks to get things working than with WS on and ISA bus heavily overclocked)

Topbench seems to favor the DLC as well.

Reply 307 of 320, by nzoomed

User metadata
Rank Member
Rank
Member

I wish I still had my 8433UUD! I sold it off about 20 years ago now 🤣

Anyway, as far as fastest 486 goes, doed overclocking count?
One of my mates at tech bought in his overclocked 486 system into class.
It was overclocked to 200 MHz!
He had to change the crystal on the motherboard from memory. I think it was an AMD-DX4100 that he overclocked. Dont remember the model of the motherboard. This was about 17 years ago.

Reply 308 of 320, by ph4nt0m

User metadata
Rank Member
Rank
Member
nzoomed wrote on 2020-08-10, 00:58:
I wish I still had my 8433UUD! I sold it off about 20 years ago now lol […]
Show full quote

I wish I still had my 8433UUD! I sold it off about 20 years ago now 🤣

Anyway, as far as fastest 486 goes, doed overclocking count?
One of my mates at tech bought in his overclocked 486 system into class.
It was overclocked to 200 MHz!
He had to change the crystal on the motherboard from memory. I think it was an AMD-DX4100 that he overclocked. Dont remember the model of the motherboard. This was about 17 years ago.

No way a DX4 could do 200MHz unless it was a relabelled 5x86. Even then, there is a difference between booting to DOS and running some simple benchmarks showing those 200MHz vs. actually doing 200MHz without errors in any environment.

My Active Sales on CPU-World

Reply 309 of 320, by nzoomed

User metadata
Rank Member
Rank
Member
ph4nt0m wrote on 2020-08-13, 19:53:
nzoomed wrote on 2020-08-10, 00:58:
I wish I still had my 8433UUD! I sold it off about 20 years ago now lol […]
Show full quote

I wish I still had my 8433UUD! I sold it off about 20 years ago now 🤣

Anyway, as far as fastest 486 goes, doed overclocking count?
One of my mates at tech bought in his overclocked 486 system into class.
It was overclocked to 200 MHz!
He had to change the crystal on the motherboard from memory. I think it was an AMD-DX4100 that he overclocked. Dont remember the model of the motherboard. This was about 17 years ago.

No way a DX4 could do 200MHz unless it was a relabelled 5x86. Even then, there is a difference between booting to DOS and running some simple benchmarks showing those 200MHz vs. actually doing 200MHz without errors in any environment.

All I can say is I saw it with my very own eyes and it definitely was an AMD-DX4.
He had to change a crystal on the motherboard to increase the clock speed.

It ran windows 95 fine too.

I should track the guy down and see if he still has it!

Reply 311 of 320, by nzoomed

User metadata
Rank Member
Rank
Member
The Serpent Rider wrote on 2020-08-13, 22:32:

AMD DX4 could be anything. What matters is part number.

Im pretty sure it was only an AM486 DX4-100, perhaps it was a DX4-120 which would have been more suitable for overclocking, but either way it definitely was a 486 and he showed us all the chip with the heatsink removed.
Im going to try and contact the guy to see if he still has the machine. This was way back in 2003 and he had overclocked it some years back even then, but had kept it to brag to everyone 🤣

Are you suggesting that some of the later AM486 chips were really a 586?
a 4x multiplier at 50MHz bus would gove you 200MHz quite easily.

Reply 313 of 320, by Anonymous Coward

User metadata
Rank l33t
Rank
l33t

5x86 is indeed a 486, but with minor differences. The very last of AMD DX/4 chips had the full 16kb internal cache like the X5 as well as the 3x/4x multipliers. They were also .35u.
200MHz is a pretty long shot even a 5x86. It can be done with L1 cache disabled, but that makes it somewhat slow. I have yet to see an example with working L1 cache.

The ST486DX4 was also made on .35u process. Has anyone managed to get 150MHz out of it?
TI486DX4 was .42u and said to be not terribly overclockable.
I'm still not sure about what IBM used on their DX4, anyone know?

"Will the highways on the internets become more few?" -Gee Dubya
V'Ger XT|Upgraded AT|Ultimate 386|Super VL/EISA 486|SMP VL/EISA Pentium

Reply 314 of 320, by nzoomed

User metadata
Rank Member
Rank
Member
Anonymous Coward wrote on 2020-08-14, 01:36:
5x86 is indeed a 486, but with minor differences. The very last of AMD DX/4 chips had the full 16kb internal cache like the X5 a […]
Show full quote

5x86 is indeed a 486, but with minor differences. The very last of AMD DX/4 chips had the full 16kb internal cache like the X5 as well as the 3x/4x multipliers. They were also .35u.
200MHz is a pretty long shot even a 5x86. It can be done with L1 cache disabled, but that makes it somewhat slow. I have yet to see an example with working L1 cache.

The ST486DX4 was also made on .35u process. Has anyone managed to get 150MHz out of it?
TI486DX4 was .42u and said to be not terribly overclockable.
I'm still not sure about what IBM used on their DX4, anyone know?

I have no idea what else he tweaked to make it work, but he certainly would have pushed the limits to get it working at that clock speed.
I always thought a 5x86 was sort of half way between a 486 and a pentium with basically a few extra instructions/extensions added to the set?

Reply 316 of 320, by Anonymous Coward

User metadata
Rank l33t
Rank
l33t

5x86 does have some 5th (or even 6th) generation features, and has a pretty interesting architecture, but from a performance point of view it is much closer to a 486 than to a Pentium. It also uses the 486 bus. I'd call the Cyrix 5x86 something like a 486+.

"Will the highways on the internets become more few?" -Gee Dubya
V'Ger XT|Upgraded AT|Ultimate 386|Super VL/EISA 486|SMP VL/EISA Pentium

Reply 317 of 320, by The Serpent Rider

User metadata
Rank l33t
Rank
l33t

After doing my homework, so to speak, I think I've finally know how to push 486 over 20 fps in Quake.

Get up, come on get down with the sickness
Open up your hate, and let it flow into me

Reply 318 of 320, by The Serpent Rider

User metadata
Rank l33t
Rank
l33t

Success! 20.2 fps on overclocked 486!

Quake 20 fps on AMD 5x86-180 Mhz.jpg
Filename
Quake 20 fps on AMD 5x86-180 Mhz.jpg
File size
547.88 KiB
Views
135 views
File license
Public domain

Get up, come on get down with the sickness
Open up your hate, and let it flow into me

Reply 319 of 320, by Intel486dx33

User metadata
Rank l33t
Rank
l33t

Have you had any success over clocking the AMD 5x86-p75-133 ADZ over the AMD-p75-133-ADW ?
I have been able to over clock the “ADW” to 160mhz fine.

How high have you been able to over lock the “ADZ” variant ?

Wikipedia posts that the “ADZ” variant is better suited for over clocking because it can endure higher temps.

But I have not read about anyone going over 160mhz ?