I ran some tests with my TX97 at 83MHz FSB: The computer starts o.k. with 83MHz x5.5 multiplier (458MHZ for my K6-III) but some games crash or cause GPFs. These are mostly Glide games on my Voodoo3 (Unreal for example). Interestingly my PowerVR PCX2 had no problems at all so far.
I then reduced the multiplier to 5.0 (415MHz) and things became stable again. As I did not lower the FSB, I suspect that the board's voltage regulators cannot provide enough power to drive the K6 at 450MHz *and* a 83MHz FSB.
So far I did not test with 256MB RAM because I'd have to rip apart my whole system in order to get access to the DIMM slots.

pyrogx wrote:

So far I did not test with 256MB RAM because I'd have to rip apart my whole system in order to get access to the DIMM slots.

I'd be interested to know if your 430TX system will even boot with 256MB RAM (128MB x 2) and 83 MHz FSB. Mine would not. It needs 1 stick of RAM to even turn on at 83 MHz.

pyrogx wrote:

I ran some tests with my TX97 at 83MHz FSB: The computer starts o.k. with 83MHz x5.5 multiplier (458MHZ for my K6-III) but some games crash or cause GPFs. These are mostly Glide games on my Voodoo3 (Unreal for example). Interestingly my PowerVR PCX2 had no problems at all so far.
I then reduced the multiplier to 5.0 (415MHz) and things became stable again.

Was it repeatable? My symptoms of crashing were very unpredictable and seldom occured. Which K6-III CPU are you refering to in this test?

I 'stumbled upon' this interesting page and I remembered the mobo is mentioned several times in this topic (ASUS TX97-E). He's using a PCI Banshee with the K6-2 at ~500 (on 83MHz bus), PC-100 SDRAM etc. Read on here:
http://www.lairdslair.com/TX97E.shtml

P.S.

also from the link on that page ("CPU Upgrade: Getting the AMD K6-2+ / K6-III+ to work on your Super Socket 7 board" by Jan Steunebrink -- includes 430TX boards):
http://web.inter.nl.net/hcc/J.Steunebrink/k6plus.htm

And, no I don't have a 430TX board myself at this moment 🙁

I'd be interested to know if your 430TX system will even boot with 256MB RAM (128MB x 2) and 83 MHz FSB. Mine would not. It needs 1 stick of RAM to even turn on at 83 MHz.

I put in a second 128MB DIMM module and ran some tests. The system starts without problems (K6-III/400AHX, family 5 stepping 9 model 1, 83MHz, x5.0 ==> 415MHz), but only recognizes 192MB, probably because the memory sticks aren't identical.
But apart from that the system seems to be stable: I ran Unreal without crashes, ran memtest for two hours without any errors reported. I also successfully compiled a Linux kernel and ran some OpenGL applications under Linux. This is usually a "safe" way to test if your system has a problem: If a kernel compilation stops with the compiler segfaulting, there is most certainly something wrong.

Was it repeatable? My symptoms of crashing were very unpredictable and seldom occured.

Yes, more or less. In Unreal it did not crash immediately and never at the same spot, but the crash occured within a few minutes of gameplay.

pyrogx wrote:

I put in a second 128MB DIMM module and ran some tests. The system starts without problems (K6-III/400AHX, family 5 stepping 9 model 1, 83MHz, x5.0 ==> 415MHz), but only recognizes 192MB, probably because the memory sticks aren't identical.

That sounds like your second stick of RAM is of a too high density for your board.
I'd bet that if you put in just that one stick of 128mb, your system will only see 64mb 😉

QDI Titanium T1B+:

This old board can use 83MHz FSB and is quite stable at that frequency. It accepts 256MB DIMMs too. 😀

DrSwizz wrote:

This old board can use 83MHz FSB and is quite stable at that frequency. It accepts 256MB DIMMs too. :-)

Could you test your system with 2 sticks of 128 MB and 83 MHz? There is a caviat in the 430tx manual about using a single 256 MB module. Is the stick being treated as x4? Remember, double sided DIMMS constintute 2 rows, as if you had inserted 2 single-sided modules.

feipoa wrote:

Did the motherboard die?
Were the core and I/O switching regulators getting hot enough to need heatsinks? I see from images online that your board didn't use any heatsinks on the regulators.

Yep, I burned the board + CPU later while trying to get 500 MHz from that K6-III. The regulators got very hot even at 450 MHz, that's why i mounted heatsinks in the first place.

@feipoa:

Sorry, I already have disassembled the parts I put together to make the screenshot. I am fairly sure that I have used 2 x 128MB modules in the past though.
I don't know how the single 256MB module is treated and I do not know how to determine that either.

QDI Titanium T1B+:

This old board can use 83MHz FSB and is quite stable at that frequency. It accepts 256MB DIMMs too.

I have one of these things in my pile of boards. I got it from a good friend who bought it new. I seem to recall a few issues. Speedeasy was a pain in the ass to work with (it never seemed to hold my settings) and I believe there were RTC issues. It kept losing time even with new batteries. I believe my friend RMAed it a few times but never got one that didn't have the problem.

Well , I am now running my trusty qdi titanium iiib to 6*83 aka 500mhz with my AMD k6-2 300 cxt core oc to 500mh on 2.3v ! The PC is super steady yet I use an Adaptec scsi 2940u controller and hard disk.

Rock solid at 2.3vcore

My super cpu.My original k6-2 400mhz cxt core could only reach up to 450.475 was a no post.

Nice results.

Must have been this thread I was remembering when I was asking this the other month Informal 430TX board survey in comparison with other Intel Triton, do they do 83Mhz (or better) as well/often???

Still, although there are a number of 430TX that do 83 reliably, I am regarding statements like "All 430TX should do 83" as the same level of BS as "All 440BX should do 133"

Time for AMD k6-3+ 500mhz on 430tx.

This post is a real blast from the past. I run my K6-3+ system at 6x83= 500 MHz without issue. I recall I had to remove the onboard L2 TAG RAM to get it working well.

MSI 5148 PMMX 166 clocked at 262 MHz 75*3.5, 64mb SDRAM PC100 and a modern Rage XL PCI 8mb, with a modern Maestro 2SC PCI, With the tightest timings possible on the BIOS; no problems so far, rock solid stability for months after the replacement of the SBawe64 that died suddenly. But I never had any problems with the 75Mhz bus in this board.
Note, with modern, I'm referring to cards bought from AE.

Running 75x3 on a P200MMX on an Abit PX5 (TX Chipset) but remembering the 90's and having 2 DeathStars on the IDE controller, 83 MHz is something I don't dare trying 😅