http://users.erols.com/chare/elec_pentium.htm
The 3.3V Version is rated at 9.75A, the 7.5A of the 200 are for split-voltage 2.9/3.3V.
I'd stick with 200 MHz to be on the safe side, at max I'd use 75x3 (if the board supports 75 and is stable)
K6 233(3.2/3.3v) on single voltage mb(3.3v) - award bios
-
- The BIOS Patcher
- Posts: 81
- Joined: Mon May 20, 2002 9:20 pm
- Location: The Netherlands
- Contact:
Just to keep you all updated on the BIOS issue, Space Marine is now running the REV:F10 BIOS. This is the latest official BIOS I could find for the Lucky Star P54CE Ver.F board and it is a close match with the original REV:F03 BIOS he send me.
This 02/14/97 REV:F10 BIOS has full support for the 3.2V/ 3.3V K6, but not for the later 2.2V K6 and K6-2 models.
Greetings,
Jan.
This 02/14/97 REV:F10 BIOS has full support for the 3.2V/ 3.3V K6, but not for the later 2.2V K6 and K6-2 models.
Greetings,
Jan.
The Unofficial AMD K6-2+ / K6-III+ page
http://web.inter.nl.net/hcc/J.Steunebrink/k6plus.htm
http://web.inter.nl.net/hcc/J.Steunebrink/k6plus.htm
hey Jan, nice to see you posting here :) it's been a while..
support for the later 2.2v K6/K6-2 would require split voltage on the board. since it doesn't support split voltage out of the box and doesn't even feature a vrm slot there's no use in supporting those processors on a board like this anyway.
support for the later 2.2v K6/K6-2 would require split voltage on the board. since it doesn't support split voltage out of the box and doesn't even feature a vrm slot there's no use in supporting those processors on a board like this anyway.
-
- The New Guy
- Posts: 1451
- Joined: Fri Mar 29, 2002 10:32 pm
- Location: Pennsylvania, USA
<- concurs with cp.
Hi Jan!
Hi Jan!
CPU - DFI 586IPVG, K6-2/+ 450 (Cyrix MII 433), i430VX, 128MB EDO.
BIOS patched by BiosMan (Jan Steunebrink).
BIOS patched by BiosMan (Jan Steunebrink).
-
- BIOS Rookie
- Posts: 39
- Joined: Sun Nov 26, 2006 2:17 am
My motherboard is dead!
I installed correctly the k6, it started but no POST done.
Then i returned to my p133 but no post too.
Maybe that mainboard was completely incompatible with k6.
Any idea or i have to launch it out of window?
I installed correctly the k6, it started but no POST done.
Then i returned to my p133 but no post too.
Maybe that mainboard was completely incompatible with k6.
Any idea or i have to launch it out of window?
-
- BIOS Rookie
- Posts: 39
- Joined: Sun Nov 26, 2006 2:17 am
Today happened a strange thing:
i tryed another time to start the mb with 133 and it starts, and now it is running.
Onestly i don't know what happened, but i didn't changed nothing from when i reinstalled it to today.
However a moderate fan was installed, not directly on cpu and VRM where i use the old ones, but at about 5 centimetres.
I don't know if is a good idea to risk again on mounting the k6.
i tryed another time to start the mb with 133 and it starts, and now it is running.
Onestly i don't know what happened, but i didn't changed nothing from when i reinstalled it to today.
However a moderate fan was installed, not directly on cpu and VRM where i use the old ones, but at about 5 centimetres.
I don't know if is a good idea to risk again on mounting the k6.
-
- BIOS Rookie
- Posts: 39
- Joined: Sun Nov 26, 2006 2:17 am
Jumpers was correct but i didnt clear the cmos.edwin wrote:you have reset the jumpers properly? You did a cmos clear?
Mb doesnt have an erasing jumper, for clearing cmos have i only to remove cmos battery?
-
- The Hardware Archivist
- Posts: 6286
- Joined: Wed Mar 20, 2002 7:11 pm
- Location: Netherlands
- Contact:
yeah and disconnect from mains power to be sure.
the VRM may have indeed overheated, it might be a good idea to put a large heatfinger on the power transistor on that and have a larger fan blow over it...
the VRM may have indeed overheated, it might be a good idea to put a large heatfinger on the power transistor on that and have a larger fan blow over it...
edwin/evasive
Do not assume anything
System error, strike any user to continue...
Do not assume anything
System error, strike any user to continue...
-
- BIOS Rookie
- Posts: 39
- Joined: Sun Nov 26, 2006 2:17 am
I don't thing that it gone overheated, i touched it just after the last trying (when it doesnt post) and was cold, like metal at ambient temperature.
-
- BIOS Rookie
- Posts: 39
- Joined: Sun Nov 26, 2006 2:17 am
I'm posting from the k6! It's up and running since 3 hours!
The problem probably was that i hadn't cleared cmos.
I have installed old cpu fan on vrm and used an old gf2mx fan as cpu fan.
The case is opened: i can touch cpu heatsink and is still cold
Vrm airflow seems a bit hotter then cpu airflow, but the difference is barely noticable, and in general is very little bit hotter then ambient temperature.
I'm thinking that probably it will not suffer much on setting cpu at 233 (original k6 speed) instead of 200.
However i have another question:
win95 recognize cpu as a general pentium, and not a k6 mmx (bios recognization is ok).
Does it influence performances much? Does any k6 mmx patch for win95a exist?
I'm not noticing big performance improvements from old p133.
The problem probably was that i hadn't cleared cmos.
I have installed old cpu fan on vrm and used an old gf2mx fan as cpu fan.
The case is opened: i can touch cpu heatsink and is still cold
Vrm airflow seems a bit hotter then cpu airflow, but the difference is barely noticable, and in general is very little bit hotter then ambient temperature.
I'm thinking that probably it will not suffer much on setting cpu at 233 (original k6 speed) instead of 200.
However i have another question:
win95 recognize cpu as a general pentium, and not a k6 mmx (bios recognization is ok).
Does it influence performances much? Does any k6 mmx patch for win95a exist?
I'm not noticing big performance improvements from old p133.
-
- BIOS Guru
- Posts: 3153
- Joined: Thu Mar 21, 2002 8:16 pm
- Location: Near Hannover (CEBIT) Germany
- Contact:
Use Everest or Sandra to check for K6 featrures been enable or not. Especially this should be MMX and Write Allocation.
It is way faster than your old CPU not only because of MHz but also 'cause of quadrupled L1-Cache. (2x8KiB on P1-133, 2x32 on K6)
It is way faster than your old CPU not only because of MHz but also 'cause of quadrupled L1-Cache. (2x8KiB on P1-133, 2x32 on K6)
-
- The New Guy
- Posts: 1451
- Joined: Fri Mar 29, 2002 10:32 pm
- Location: Pennsylvania, USA
If you use Everest, please post a dump of the CPU MSR's.
To do this, right-click on the bottom bar, select CPU Debug, then K6 MSR Dump.
Thanks.
To do this, right-click on the bottom bar, select CPU Debug, then K6 MSR Dump.
Thanks.
CPU - DFI 586IPVG, K6-2/+ 450 (Cyrix MII 433), i430VX, 128MB EDO.
BIOS patched by BiosMan (Jan Steunebrink).
BIOS patched by BiosMan (Jan Steunebrink).
-
- BIOS Rookie
- Posts: 39
- Joined: Sun Nov 26, 2006 2:17 am
I'm using everest
- MSR dump is empty
- this is cpuid+msr dump
- The cpu is recognized as K6 at 200mhz (66x3, is right) and both mmx and L1 cache are ok
- everest seems unable to say if write allocation is active or not. Havent find that option written
BUT, there is a problem:
- L2 cache is not recognized
I have 256k of pipelined burst cache soldered onto motherboard.
But it has a 430VX and im using 128mb of ram.
VX support only 64 mb of ram for caching, so maybe is ram ammount the problem.
In your opinion, is better having 128mb of ram without L2 cache or 64mb using 256k (plus an eventual coast module) of L2?
- MSR dump is empty
- this is cpuid+msr dump
------[ Logical CPU #0 ]------
allcpu: Valid
CPUID 00000000: 00000001-68747541-444D4163-69746E65
CPUID 00000001: 00000562-00000000-00000000-008001BF
CPUID 80000000: 80000005-00000000-00000000-00000000
CPUID 80000001: 00000662-00000000-00000000-008005BF
CPUID 80000002: 2D444D41-6D74364B-202F7720-746C756D
CPUID 80000003: 64656D69-65206169-6E657478-6E6F6973
CPUID 80000004: 00000073-00000000-00000000-00000000
CPUID 80000005: 00000000-02800140-20020220-20020220
------[ MSR Registers ]------
PerformanceFrequency = 1193180
- The cpu is recognized as K6 at 200mhz (66x3, is right) and both mmx and L1 cache are ok
- everest seems unable to say if write allocation is active or not. Havent find that option written
BUT, there is a problem:
- L2 cache is not recognized
I have 256k of pipelined burst cache soldered onto motherboard.
But it has a 430VX and im using 128mb of ram.
VX support only 64 mb of ram for caching, so maybe is ram ammount the problem.
In your opinion, is better having 128mb of ram without L2 cache or 64mb using 256k (plus an eventual coast module) of L2?