Babasha wrote on 2023-11-01, 17:54:Nice 😉
Kingston Turbochip TC366 with AMD K6-2 366MHz
Wow! Didn’t even know such upgrades existed. Is that s7 drop in upgrade?
Babasha wrote on 2023-11-01, 17:54:Nice 😉
Kingston Turbochip TC366 with AMD K6-2 366MHz
I have an Evergreen Spectra that came with a K6-2 400 - I replaced the CPU with a K6-3+ and I have it installed on a PCChips M520. My plan is to build a Necroware voltage regulator and use it instead.
"We do these things not because they are easy, but because we thought they would be easy."
CharlieFoxtrot wrote on 2023-11-01, 17:40:Yeah, I’m sure. Clock for clock kt266a did beat 760 and sometimes in significant margin. KT266a had especially much better memor […]
BitWrangler wrote on 2023-11-01, 17:04:CharlieFoxtrot wrote on 2023-11-01, 16:56:It sure isn't the fastest chipset,
You are sure you are sure? There is a set of CMOS optimisations and WCPRedit files that can make these things scream, and when I say scream, I mean top out somewhere north of 170FSB leaving KT333 in the dust and giving nforce a run for it's money. OPPainter was the main guy behind these tweaks. It was hard to dig out info before, might be even harder since in the past year google AI enhanced it's search to make it stupider.
Yeah, I’m sure. Clock for clock kt266a did beat 760 and sometimes in significant margin. KT266a had especially much better memory performance.
I wouldn’t even bother comparing it to newer chipsets that came for socket A, there is just no contest. It is not a surprise that 760 faded away quite quickly late 2001 / early 2002.
It wasn’t particularly good overclocker either. Many kt266a OC boards could reach the same numbers and best ones like EpoX 8kha+ got 1/5 and 1/6 PCI dividers and could reach 200+MHz, best results were around 220. Many KT333 boards went just fine to the 215-220 mark.
KG7 was probably the best 760 board released, but it got obsolete quite fast after the release mainly due better kt266a chipset boards.
I am talking synchronous with CPU FSB though, not merely DDR speeds.
Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.
BitWrangler wrote on 2023-11-01, 19:59:I am talking synchronous with CPU FSB though, not merely DDR speeds.
Sure, me too. There is very little point in running socket A other than 1:1 FSB and memory. In fact, there is quite a performance hit with running them asynchronously.
Well I've only ever seen CpuZ suicide screenies of KT333 much over 180 and never any runs of anything like 3Dmark or super pi 1M or bigger at better than 170 to confirm it doesn't crash instantly. Best KT333 speed I can find on HWBot at the moment is only 167 for a 3Dmark run. Granted there's not many 761 on there either but it just missed being a thing until after everyone thought they upgraded to KT266. Really annoying amount of "bit rot" though with blocked photobucket, non existant imageshack, tinypic etc and all the other dead image sites that ppl put their screenies on. I just realized that mine were probably on imageshack if they were ever uploaded also so no point looking.
Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.
I ran my Epox 8K3A+ at 209 FSB for a long time. I pushed it up to 215, and it was still stable, but my hard drive speed fell off sharply at anything over 209 so I went back to that. I upgraded to an 8RDA3+ because the 8K3A+ developed a case of bad capacitors.
"We do these things not because they are easy, but because we thought they would be easy."
Okay, yah, I have been mistaken, I must have thought Epox Alphabetsoup models were Nforce boards.
Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.
Got these today：
1. Matrox Millennium II, 8MB of WRAM is integrated on board.
2.MSI MS-4131G, 128K L2 cache installed,The location of the CPU socket is peculiar.
Am386DX-40, 386-VC-H, 16MB, GD5429, ES1868F
Am486DX5-133, HIPPO-15, 32MB, Riva 128, SB16
K6-III+550, VA-503+, 128MB, Voodoo3 2000, AWE32
Tualatin-1.4G, 694X, 512MB, G400, Voodoo2, AWE64
BitWrangler wrote on 2023-11-01, 22:06:Well I've only ever seen CpuZ suicide screenies of KT333 much over 180 and never any runs of anything like 3Dmark or super pi 1M or bigger at better than 170 to confirm it doesn't crash instantly. Best KT333 speed I can find on HWBot at the moment is only 167 for a 3Dmark run. Granted there's not many 761 on there either but it just missed being a thing until after everyone thought they upgraded to KT266. Really annoying amount of "bit rot" though with blocked photobucket, non existant imageshack, tinypic etc and all the other dead image sites that ppl put their screenies on. I just realized that mine were probably on imageshack if they were ever uploaded also so no point looking.
I ran my EpoX 8k5a2 kt333 board at 210 back in the day with no problems and that certainly wasn’t anything exceptional. I was limited by my memory configuration as I had mixed ram sticks with different chips, so memory OC was bit janky, although both of my sticks were fine performing memory itself.
Yeah, there is not that much data easily available from those days, but some old forum discussions do exist, as well as reviews. Socket A is extremely fascinating platform in the sense that the ”chipset wars” during those days introduced huge amount of chipsets (and boards) with constantly improving performance and features during a relatively short period of time and it even pushed AMD out of the chipset market.
I certainly remember good 200/400 DDR being a problem, for a couple of years, and was just coming right as DDR2 came in. Anyway, my point was really that 761 had a second lease of life when everyone had moved to the new shiny, and there is an unbelievable performance improvement possible. Maybe I was thinking first wave of KT333 was a dud tho, due to there being many early boards where they just slapped the KT333 in place of the KT266A, so really they only did marginally better than the board did with a KT266A. Though I might have discontinuous recollections due to paying attention for a couple of months, then dealing with life for a few months, then paying attention then not etc.
Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.
zwrr wrote on 2023-11-02, 03:43:Got these today： 1. Matrox Millennium II, 8MB of WRAM is integrated on board. 2.MSI MS-4131G, 128K L2 cache installed,The locati […]
Got these today：
1. Matrox Millennium II, 8MB of WRAM is integrated on board.
2.MSI MS-4131G, 128K L2 cache installed,The location of the CPU socket is peculiar.MGA-1.jpg
MGA-2.jpg
MS4131G-1.jpg
MS4131G-2.jpg
Nice one - I had quite a few from these.
At first I couldn't believe that these really carried 8MiB.
BitWrangler wrote on 2023-10-30, 22:11:I shouldn't go to look stuff up on fleabay... because then what happens is I see something shiny... in this case a ULSI US83C87 FPU DX/DLC-40MHz .. while they're cheap before everyone notices that a youtube dude has been clocking them over 50mhz. Slow boat from China though... good thing I'm not in a hurry for it. Hoping it lets me run my BL3 system at 2x50.
It's on it's way, but I just found out I had no justification for buying it... the board with a 386 socket I thought I could put the "spare" FPU in after I'm done playing on the BL3 machine and it keeps the best, doesn't have a 387 socket, doh. This is bad, I am obviously now at some point going to use the existence of a spare FPU to justify buying a board I can put it into.... possibly... I dunno, I'm still finding crap I forgot about, maybe do have another board, so will have to hold that thought until I'm 100% positive I've checked all nooks and crannies... which will take a while still.... so yeah no rush 🤣
Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.
update , I buyed sunday a (partially)broken Quadro K4200 , 4gb gddr5 , it gave a code 43 on the gpu under windows , and some horiz lines(after driver install) ....
Cleaned the whole thing , after dissasambly , then hot ironed evrything (gpu+ram chips) but had no effect , no progress was made.
Eventually I managed to find the bios and Kepler bin editor , so I used that and just keep lowering the frequencies till it was stable enough to not give code43. (rewrote the firmware/gpu/bios/bin 4-5 times)....
I got it to a point I can load CS:S (HL2 engine) , at 640x480 or 1024x768 ... but in other 3d it just gives me , blank screen for a moment, and sometimes it starts most of the times it goes out of the 3dx app...
- I am stuck , not sure what else to do , voltage is back to stock (in .bin file , and I gave a uper 0.025v room in case it needed)
How do I check if it's a ram module? or the GPU ? causing evrything? (to narrow down the problem , it also has some laptop style VRM/MOSFET's - would one of them be actingup?
Yeah if it was a CPU I'd say try it on another motherboard, if no symptoms there, then the mosfets and power circuits are shot, if still as bad the CPU has got electromigration deterioration full set in from overheat, overvolt or overclock or all three. But not so easy just to swap the GPU chip around so up to you whether you wanna rebuild the power and cross your fingers or call this one a lost cause.
Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.
DerBaum wrote on 2023-10-08, 21:53:...
Now the really bad news.
I think i have killed the 5 1/4 floppy drive...
Because i dont feel confident in finding the failure in the drive i destroyed... i bought another one.
I bought a very similar model so i can swap the faceplate to the brown one my drive has on it.
The drive i got is untested... wich is not a good sign in general... but it was the cheapest one for 30 euros.
Who the hell decided that a used untested floppy drive is now 30 bucks? depressing...
At least it looks clean...
I think today or tomorrow i have time to test it.
FCKGW-RHQQ2
DerBaum wrote on 2023-11-02, 16:53:Because i dont feel confident in finding the failure in the drive i destroyed... i bought another one. […]
DerBaum wrote on 2023-10-08, 21:53:...
Now the really bad news.
I think i have killed the 5 1/4 floppy drive...Because i dont feel confident in finding the failure in the drive i destroyed... i bought another one.
I bought a very similar model so i can swap the faceplate to the brown one my drive has on it.
The drive i got is untested... wich is not a good sign in general... but it was the cheapest one for 30 euros.
Who the hell decided that a used untested floppy drive is now 30 bucks? depressing...
At least it looks clean...
I think today or tomorrow i have time to test it.
2023-11-02 14.48.30.jpg2023-11-02 14.47.40.jpg2023-11-02 14.46.38.jpg
5.25” floppy drives seem to be quite insanely priced nowadays, although occasionally you can bump into cheap ones. DD drives seem to be even pricier.
I have couple 55GFR drives myself and in my experience they are reliable and very flexible drives as they can be easily ”locked” for 300rpm if needed.
55GFR series was also quite late model and yours is the final revision and I’d guess it is made in 92 or 93 being one of the last manufactured 5.25 units (at least from Teac). This means that very likely it has seen quite little use and mechanically it is in great condition. I have one 7193-U and it is manufacterd in 1993.
CharlieFoxtrot wrote on 2023-11-02, 18:36:...
55GFR series was also quite late model and yours is the final revision and I’d guess it is made in 92 or 93 being one of the last manufactured 5.25 units (at least from Teac). This means that very likely it has seen quite little use and mechanically it is in great condition. I have one 7193-U and it is manufacterd in 1993.
the one i presumingly broke (by reversing the data cable) was a 55GFR-571-U. It is way more complicated than the one that came today.
I really want to get it working again but have to get better diagnosing it.
FCKGW-RHQQ2
DerBaum wrote on 2023-11-02, 18:49:CharlieFoxtrot wrote on 2023-11-02, 18:36:...
55GFR series was also quite late model and yours is the final revision and I’d guess it is made in 92 or 93 being one of the last manufactured 5.25 units (at least from Teac). This means that very likely it has seen quite little use and mechanically it is in great condition. I have one 7193-U and it is manufacterd in 1993.the one i presumingly broke (by reversing the data cable) was a 55GFR-571-U. It is way more complicated than the one that came today.
I really want to get it working again but have to get better diagnosing it.
I’m quite surprised that flipping the data cable actually broke something on the drive, but I guess it is possible. It is not like there is some large voltage or current on that cable and I have managed to do that to 3.5” drives before and it hasn’t caused any long term issues.
It did used to happen that flipping the cable could kill things... but I remember more killed controllers in general than drives.
Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.
I am also guilty of reversing a 3.5" floppy cable, and there was no damage. I've never reversed a 5.25" floppy connector. Aren't those card edges keyed?
