Shodan486 wrote:

I think now I can enjoy the fruit from KT133. This is what I call fun !!! I'll try to post some pics when time found.

Awesome signature! If you have time, please run 3DMark 2001 s.e. (and also the 2000 version), and post your scores. Thanks.

ux-3 wrote:

I have decides to join the fun and have ordered my first ever KT133A

What mobo did you choose? It would be interesting to learn if you have any problems with your AWE32. Is it a PnP CT3980? I tried a non-PnP CT3900 in an Abit KT133A based mobo, and it made crackling noises. I switched to an AWE64, and it worked fine.

BTW, I really am keen on recapping my Epox board. I may be able to afford a solder station + bits required in about 2 months time..

retro games 100 wrote:

ux-3 wrote:

I have decides to join the fun and have ordered my first ever KT133A

What mobo did you choose? It would be interesting to learn if you have any problems with your AWE32. Is it a PnP CT3980? I tried a non-PnP CT3900 in an Abit KT133A based mobo, and it made crackling noises. I switched to an AWE64, and it worked fine.

I didn' really choose a board. I decided that for experimenting, I don't want to use my XP-M. You do know why! So I bought something that came "ready to fry" except for a power supply. I was looking for ISA, KT133a, "functioning" and paypal. When its here and powered up without smoke, I start making plans. It is too early yet.

After a day, I came to realise why I hated my old SB32. When installing this card, it took forever, and each new attempt brought about new reactions from diagnose.exe. Finally, it discovered the RAM, the different music options etc. I remembered then that I got there once before, roughly 15 years ago.

But the card still isn't making me happy: Archimedean Dynasty just didn't run. When I swapped back the old SB16, it worked just fine. So I decided that I don't want another source of trouble with 4 to boot install progs in order to get even simple SB compatibility. I was intrigued because this AWE could also take a daughterbord. I figured I could get a SB16 plus AWE but it doesn't work that way. Also CHoff crashes the machine when the AWE is installed.

I have two daughterboards and another CT-1740 enroute and will check the entire issue again, once the stuff is here.

The reason I got no AWE64 was simply that there is no version with the DB-connector. Once you consider an AWE64, why not use a SBLive! ? What options would the AWE64 have, that the Live! would not?

You have two DBs enroute? They're not that common anymore. You must be "hunting high and low" for this stuff. Hehe.

Yes, I now understand why you do not want the AWE64, because there is no DB connector on it. Re: CT1740 - I can't remember if that model has the stuck/hanging MIDI bug. If it doesn't, and it's OK, then it will give you "OK" sound quality. BTW, I've had good success with the CT2230 models. They sound better than the CT17xx series, and I really don't remember having any stuck/hanging MIDI bug problems with them.

>> Once you consider an AWE64, why not use a SBLive! ? What options would the AWE64 have, that the Live! would not?

The AWE64 is useful, particularly on the KT133A boards. I noticed when I used the AWE64 with a KT133A board, I had no problems with crackles, odd sound delays, latency problems, etc. It just worked. With the SBLive!, I think old usenet postings from that era in computing history contain comments about PCI latency problems with the SBLive! card.

Vogons user prophase_j has a "retro rocket" system with an AWE64 in the ISA slot, and an AU8830 chip-based Diamond MX300 sound card in one of the PCI slots. That sound card can host a daughterboard. (I'm not too sure if it works in "pure DOS" though.) I once tried a Montego 2 PCI sound card with a daughterboard. The DB works in pure DOS - it requires a special DOS driver to be loaded in memory for it to work.

Will do, RetroGames.

3DMark2001SE = 6483
3DMark2000 = 6879

Highest details possible set, resolution was full HD - 1920x1080

Shodan486 wrote:

3DMark2001SE = 6483
3DMark2000 = 6879

Highest details possible set, resolution was full HD - 1920x1080

Thanks a lot. Actually, I've been told to do this here on Vogons: could you please rerun them, but leave the 3DMark settings on their "default" settings. That way, you put more strain on the CPU/mobo. All the tests I have done (and reported about here on Vogons) in the last couple of months have been set to the 3DMark default settings. Thanks!

Edit: I've just double-checked the mobo written in your signature, and that board has been reviewed here -

http://www.thetechzone.com/reviews/motherboar … vp6/index.shtml

It uses the VIA Apollo Pro 133A chipset.

So da default setting:

3DMark2000 = 7364
3DMark2001 = 9411

3DMark2000 PIII mode = 5313

Sorry guys, I was doing these tests in quite a chaos and I may have swithched the results from 2000 & 2001, but I will check it out.

To RetroGames: I know I have VIA Apollo 133 on the board, so what? Not getting the point of your pointing out.

Shodan486 wrote:

To RetroGames: I know I have VIA Apollo 133 on the board, so what? Not getting the point of your pointing out.

You said earlier "I think now I can enjoy the fruit from KT133", when referring to your Apollo based signature.

My friend 😅 , VIA Apollo 133a IS KT133A.

Shodan486 wrote:

My friend 😅 , VIA Apollo 133a IS KT133A.

If I am not mistaken, you are mistaken... - 😁

The Via Apollo 133a is a P3 chipset, the KT133a is an Athlon chipset.

Shodan486 wrote:

My friend 😅 , VIA Apollo 133a IS KT133A.

Actually... it isn't.

🙄

Here's an interesting web link -

http://www.tomshardware.com/reviews/shape,282.html

The title is "VIA In Shape: 6 Motherboards using the renewed KT133A Chipset"

BTW, I just received a replacement XP-M 2400+ CPU, to replace the one I accidentally killed several weeks ago. I have not unlocked it yet. It's interesting, because the 2 bridge "ends" of the L6 bridge appear to be far apart - compared to the tiny gap seen on my previous (now dead) XP-M 2400+. I am confident that I can push the tip of a sharp knife in to the 3rd bridge, in order to unlock it.

Meanwhile, I ran the Epox board using the locked XP-M, at 1800mhz. I used an ATI Radeon 8500, and got 7125 using 3DMark 2001 s.e.

Yes guys, both of you are right, I have explained myself inappropriately, thus correcting now: In all sense the KT133A should be no more nor less than the Via Apollo except the CPU support of AMD brand, which meant support of 200 & 266 MHz support mainly. I have no knowledge of any engeneering intervention towards improving the chip, it still supports only the SDRAM DIMMs - it SHOULD (please, prove me wrong) be the very Apollo, but ready for faster CPU bus. And I cannot believe it is not able to support 2GB RAM, has anyone tried it?

I know I this thread is about Athlons and KT133, but just wanted to add how strong it can be when using proper capacitors and memory modules - The CPU NEVER EVER lets you down the first, so do not underestimate its overclocking capabilities.

It is my mistake to call it KT133, Im sorry for the confusion.

Waiting for a heatsink for the chip right beside the DIMM 0 from China, soon I will reach 173Mhz, maybe more - still with the TAEPO caps 😀

Shodan486 wrote:

Red (for CPU & RAM & AGP): 1500uF 6.3v replaced by 2200uF 6.3v Green (I guess for chipset): 1000uF 16v replaced by 2200uF 6.3v ( […]
Show full quote

Red (for CPU & RAM & AGP): 1500uF 6.3v replaced by 2200uF 6.3v
Green (I guess for chipset): 1000uF 16v replaced by 2200uF 6.3v (!!!)
Purple (I guess for AGP & chipset): 1000uF 16V replaced by 1000uF 16v (better quality)

I am particularly concerned with the green couple (what is it for anyway?) and that's why I'm referring to the tradition of not replacing with lower attributes.

Regarding the green circled caps; quoted line 2 above. From my very limited understanding about recapping, I thought that you needed to have the same voltage caps, or higher voltage caps. Did you use a recapping service? Were the original caps bulged, or were the caps visibly OK but the mobo was a bit unstable? It's interesting that you are able to significantly OC the board, and all seems well with it.

We´re the same with the recapping skills, but I browsed a lot of theore and frankly, this is the very first time theory has beaten off pragmatism: I have 6 VP6 boards, one is absolutely new with the notorious Jackon caps, which you should know that there was a story of industrial espionage from, started by a fired Rubycon engineer who took some unfinished electrolytic formula information for his own purpose, guess he wanted to create his own and sell, but his team he had hired stole this info and created their own, and sold it to big manufacturers who plagued much electronics worldwide...but back to the board

The rest 5 have different caps, different capacitance values and the different withstanding voltage. I will now answer to your questions:

First, yes, you are right about the rule not to use lower voltage caps - on this particular board, only 2 caps need to be 16v - the ones on the rear edge near to the AGP slot, that is the VRM from 12v to 3.3v. The green couple circled can have lower values, it should belong to the chipset & CPUs, so using 6.3V is well enough, because it takes voltage from the 5v rail. 6.3v caps are available with much higher capacitance than the 16v, so using them is wise for overclockers. And of course, higher capacitance means lower ESR.

As for replacing, I bought all of them at ebay, so no recapping, although I have bought 6.3v 3300uF and 16v 1800uF Ultra Low ESR MBZ Rubycon series for OC improvment up to the peak of stability - these should reach the maximum limit of 178MHz FSB. As to my knowledge, there is none VP6 that has not been recapped yet, the former electrolytics failed after approx 2-3 years of normal usage, 1 1/2 year of 24h usage. It is rather rare that one has been found on ebay.

I always knew that not the chips or memory modules or the motherboards themselves are the limitation of OC, it were their electronic parts = mainly the capacitors, sometimes MOSFETs.

Found another one of these socket a boards with an isa slot, how does this one look?

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem … e=STRK:MEWAX:IT

from what I hear Iwill used to be quite good back in the day.

2Mourty wrote:

Found another one of these socket a boards with an isa slot, how does this one look?

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem … e=STRK:MEWAX:IT

from what I hear Iwill used to be quite good back in the day.

Interesting. I have tested an IWill KK266, but not that version. I talked about it here -

Two dead "retro rocket" mobos? (Half way down)

Plus-R Version 2.0 sounds quite "up to date". May be it will run a t-bred 2600+ CPU, but I cannot say for certain.

A review of the IWill KK266 Plus can be found here -

http://www.pcstats.com/articleview.cfm?articleid=801&page=1

IWill KK266 Plus R review:

http://www.overclockersonline.net/?page=artic … &num=100&pnum=0