VOGONS


First post, by Sphere478

User metadata
Rank l33t++
Rank
l33t++

Edit: solved
Here are the multiplier settings for the 2.2v 400gp:

If we assume that dip switches 5,6,7 on my freeway design motherboard correspond to bf0,bf1,bf2 respectively then:
1= on 0 = off
Bf0, bf1, bf2
? ? ? = 1x (might only work on some mobos, or may not exist on this chip)
1 0 0 = 2x
1 1 0 = 2.5x
0 1 0 = 3x
0 0 0 = 3.5x
1 0 1/ 1 1 1 = 4x

Here are the multiplier settings for a 2.9v cyrix 366gp

 1=on 0=off
(66fsb test actual mhz, not PR)
Bf0, bf1, bf2
0, 0, 0 233mhz 3.5x
1, 0, 0 133mhz 2.0x
0, 1, 0 200mhz 3.0x
1, 1, 0 166mhz 2.5x
0, 0, 1 66.7mhz 1.0x
1, 0, 1 266mhz 4.0x
0, 1, 1 66.7mhz 1.0x
1, 1, 1 266mhz 4.0x

OP:

Anyone have a cheat sheet for all the multiplier settings for a 400gp on a asus p5a?

Given that the post screen is displaying only a pr rating this is becoming difficult with the 27 possible combinations.

Last edited by Sphere478 on 2022-03-28, 13:35. Edited 20 times in total.

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 1 of 31, by waterbeesje

User metadata
Rank Oldbie
Rank
Oldbie

It's been a while since I've worked with one, but weren't those supposed to run just under 300MHz? At 95x3?

I also think you may get away with 75x4 or 83x3,5 if your board supports it.

And keep it really cool 😀

Oh just had a quick peek at the Redhill guide before I posted and they also state 95x3 as only value.

Stuck at 10MHz...

Reply 2 of 31, by Sphere478

User metadata
Rank l33t++
Rank
l33t++
waterbeesje wrote on 2021-12-24, 11:29:
It's been a while since I've worked with one, but weren't those supposed to run just under 300MHz? At 95x3? […]
Show full quote

It's been a while since I've worked with one, but weren't those supposed to run just under 300MHz? At 95x3?

I also think you may get away with 75x4 or 83x3,5 if your board supports it.

And keep it really cool 😀

Oh just had a quick peek at the Redhill guide before I posted and they also state 95x3 as only value.

Yeah, 95x3

I can get bios to report 233, 266, mmx 100mhz, 380, and that’s about it.

I found a 380 setting that is 350 mhz but it’s not stable it seems.

I’m 100fsb

Were the 400gp, 433gp multipliers all standard or is this one of those chips with weird ones?

Anyway I assume I’m at 3.5x right now but later on I’m gonna mess with the fsb and need to know how to set each multi

Edit: seems all off is 3x
That should help for right now but down the road I need to figure out the rest. I wanna try for 133fsb later

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 3 of 31, by Sphere478

User metadata
Rank l33t++
Rank
l33t++

Just updated bios. Detecting as a 433gp now w00t!

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 4 of 31, by rmay635703

User metadata
Rank Oldbie
Rank
Oldbie
Sphere478 wrote on 2021-12-24, 12:31:

Just updated bios. Detecting as a 433gp now w00t!

You have a rare chip, funny part was after the market collapsed Cyrix had a bunch of 433’s for sale, never have seen a 400

Lots of IBM 350/366 variants on the market
333 was rare, 300gp common as dirt, odd times while Cyrix was in its death throes or already dead.

They were marketing these things when 450-700mhz was common

Reply 5 of 31, by BitWrangler

User metadata
Rank l33t++
Rank
l33t++
waterbeesje wrote on 2021-12-24, 11:29:

Oh just had a quick peek at the Redhill guide before I posted and they also state 95x3 as only value.

That RedHill guide has a large number of errors and omissions, "back in the day" around 2000 when it was first up a number of enthusiasts were emailing them links and sources to correct their info, and they never did. Now those sources have gone from time and web rot and Red Hill gets cited as a source on Wikipedia *sigh*

Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.

Reply 6 of 31, by waterbeesje

User metadata
Rank Oldbie
Rank
Oldbie
BitWrangler wrote on 2021-12-24, 16:24:
waterbeesje wrote on 2021-12-24, 11:29:

Oh just had a quick peek at the Redhill guide before I posted and they also state 95x3 as only value.

That RedHill guide has a large number of errors and omissions, "back in the day" around 2000 when it was first up a number of enthusiasts were emailing them links and sources to correct their info, and they never did. Now those sources have gone from time and web rot and Red Hill gets cited as a source on Wikipedia *sigh*

Yeah, redhill does have quite some errors, and it's definitely no wiki or something... But they get things right as well. As for these fsb/multi settings I think it could be a fair starting point

Sphere478 wrote on 2021-12-24, 12:31:

Just updated bios. Detecting as a 433gp now w00t!

That's a great starter 👍

Stuck at 10MHz...

Reply 7 of 31, by Sphere478

User metadata
Rank l33t++
Rank
l33t++
rmay635703 wrote on 2021-12-24, 16:20:
You have a rare chip, funny part was after the market collapsed Cyrix had a bunch of 433’s for sale, never have seen a 400 […]
Show full quote
Sphere478 wrote on 2021-12-24, 12:31:

Just updated bios. Detecting as a 433gp now w00t!

You have a rare chip, funny part was after the market collapsed Cyrix had a bunch of 433’s for sale, never have seen a 400

Lots of IBM 350/366 variants on the market
333 was rare, 300gp common as dirt, odd times while Cyrix was in its death throes or already dead.

They were marketing these things when 450-700mhz was common

Did ibm or ST ever market 400 or 433 versions?

I’m still looking for a 433gp I’ve found three but have been too late to the party to get one

What is redhill?

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 8 of 31, by waterbeesje

User metadata
Rank Oldbie
Rank
Oldbie
Sphere478 wrote on 2021-12-24, 20:28:

(...)
What is redhill?

Since we're around the holidays I think you may have some spare time... So it may be safe to tell you 😜

https://www.redhill.net.au//ig.html

As stated, they're not always right but I think it's some pleasant reading 😀

Stuck at 10MHz...

Reply 9 of 31, by BitWrangler

User metadata
Rank l33t++
Rank
l33t++

Take it as kind of an editorial color piece a stroll through computer hardware over the years and it's fine.

Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.

Reply 10 of 31, by rmay635703

User metadata
Rank Oldbie
Rank
Oldbie
Sphere478 wrote on 2021-12-24, 20:28:
Did ibm or ST ever market 400 or 433 versions? […]
Show full quote
rmay635703 wrote on 2021-12-24, 16:20:
You have a rare chip, funny part was after the market collapsed Cyrix had a bunch of 433’s for sale, never have seen a 400 […]
Show full quote
Sphere478 wrote on 2021-12-24, 12:31:

Just updated bios. Detecting as a 433gp now w00t!

You have a rare chip, funny part was after the market collapsed Cyrix had a bunch of 433’s for sale, never have seen a 400

Lots of IBM 350/366 variants on the market
333 was rare, 300gp common as dirt, odd times while Cyrix was in its death throes or already dead.

They were marketing these things when 450-700mhz was common

Did ibm or ST ever market 400 or 433 versions?

I’m still looking for a 433gp I’ve found three but have been too late to the party to get one

What is redhill?

Via = yes

IBM/ ST Nope,
ST was still making single rail 6x86-Pr166’s into the 2000’s for washing machines and other automated gizmos but didn’t really even enter the MII era.
rumor is the real Cyrix never made any significant numbers of 366 chips either , the 366gp was an IBM creation pushing Cyrix forward since the real deal was under attrition during the National years.
IBM was the only entity pushing them forward once they were removed it wasn’t long for VIA to buyout as nothing new was being made.

Reply 11 of 31, by Sphere478

User metadata
Rank l33t++
Rank
l33t++
rmay635703 wrote on 2021-12-24, 22:14:
Via = yes […]
Show full quote
Sphere478 wrote on 2021-12-24, 20:28:
Did ibm or ST ever market 400 or 433 versions? […]
Show full quote
rmay635703 wrote on 2021-12-24, 16:20:
You have a rare chip, funny part was after the market collapsed Cyrix had a bunch of 433’s for sale, never have seen a 400 […]
Show full quote

You have a rare chip, funny part was after the market collapsed Cyrix had a bunch of 433’s for sale, never have seen a 400

Lots of IBM 350/366 variants on the market
333 was rare, 300gp common as dirt, odd times while Cyrix was in its death throes or already dead.

They were marketing these things when 450-700mhz was common

Did ibm or ST ever market 400 or 433 versions?

I’m still looking for a 433gp I’ve found three but have been too late to the party to get one

What is redhill?

Via = yes

IBM/ ST Nope,
ST was still making single rail 6x86-Pr166’s into the 2000’s for washing machines and other automated gizmos but didn’t really even enter the MII era.
rumor is the real Cyrix never made any significant numbers of 366 chips either , the 366gp was an IBM creation pushing Cyrix forward since the real deal was under attrition during the National years.
IBM was the only entity pushing them forward once they were removed it wasn’t long for VIA to buyout as nothing new was being made.

there was a via branded socket 7 cpu?

I have two 366gp chips and I see them on ebay all the time.

I’ve run across about twice as many 400s than 433s but I still l can’t seem to track down a 433 I definitely want one for my collection though. At least the 400 is basically the same thing.

Last edited by Sphere478 on 2022-01-14, 06:46. Edited 1 time in total.

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 12 of 31, by rmay635703

User metadata
Rank Oldbie
Rank
Oldbie
Sphere478 wrote on 2021-12-24, 23:13:

there was a via branded socket 7 cpu?

I have two 366gp chips and I see them on ebay all the time.

I’ve run across about twice as many 400s than 433s but I’m still l can’t seem to track down a 433 I definitely want one for my collection though. At least the 400 is basically the same thing.

Not that I know of,
My VIA made 433 still says Cyrix on it

IBM was pushed out shortly after they designed the 333/350/366gp (all same clock ?)

Also IBM was still making chips for Cyrix after they were dropped and their agreement ended post National takeover.

https://www.cnet.com/news/ibm-makes-faster-cyrix-chips/

Not sure if VIA made Cyrix chips in pr366 or not but they were very common in the USA, maybe these were national made though I never believed National went that high.

I guess if they are 2.9 volt they would have to be IBM/Cyrix, if 2.2 they are most likely done under VIA ownership

Never looked to see what the identifiers for VIA Made vrs National made were, also not sure if Cyrix chips made by IBM post breakup were labeled IBM anymore.

Reply 13 of 31, by Sphere478

User metadata
Rank l33t++
Rank
l33t++

So what multipliers did the 400gp support?

Was 3x the highest? No 3.5x or 4x?

I guess I’ll have to plug it back in and try a few settings and let everyone know

I do know that it feels really fast, like clock per clock with a k6-3+

Windows loads super fast.

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 14 of 31, by rmay635703

User metadata
Rank Oldbie
Rank
Oldbie
Sphere478 wrote on 2022-01-14, 06:47:
So what multipliers did the 400gp support? […]
Show full quote

So what multipliers did the 400gp support?

Was 3x the highest? No 3.5x or 4x?

I guess I’ll have to plug it back in and try a few settings and let everyone know

I do know that it feels really fast, like clock per clock with a k6-3+

Windows loads super fast.

There was an interesting and funny article posted long ago putting a Celeron 300a head to head with a Cyrix 433gp at the end they had to overclock the Cyrix to an Actual 433mhz to complete with the Celeron at stock speed (supposedly)

My chip supports up to 4x but I can’t say it had any multipliers past that

http://www.k6plus.com/phpBB3/viewtopic.php?t=1997

Reply 15 of 31, by BitWrangler

User metadata
Rank l33t++
Rank
l33t++

Now I can imagine that you'd have to get a Cyrix MII core up to 433 to match Quake and FPU scores of a Celeron 300a, but Integer performance is what it's spectacular at, and almost the converse should be true, that you'd have to stick a 300A on a 100mhz bus and have it at 450 before it beat a MII 433s integer scores.

Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.

Reply 16 of 31, by rmay635703

User metadata
Rank Oldbie
Rank
Oldbie
BitWrangler wrote on 2022-01-14, 14:36:

Now I can imagine that you'd have to get a Cyrix MII core up to 433 to match Quake and FPU scores of a Celeron 300a, but Integer performance is what it's spectacular at, and almost the converse should be true, that you'd have to stick a 300A on a 100mhz bus and have it at 450 before it beat a MII 433s integer scores.

I kinda figured that since their benchmark was just a couple games

What I find amusing is the MIIv (via owned) Cyrix chips all seem to easily and massively overclock just like the 300a

I tested mine at 400mhz with a meh cpu cooler and it didn’t wink and it seemed to perform very very well on my IBM Supplied SIS540 motherboard .

If Via would have taken over and released the “433” chips 2 years earlier they would have been a runaway success, bought mine brand new around 2001 for cheap out of curiosity (I still ran socket 7 in many rigs at that point)

I might have to dig maybe that thing needed a recap despite being under 3 years old, my 433gp is in storage somewhere as well

Ah the alternative universe where the National fail didn’t happen

Cyrix MII-433GP Build

Reply 17 of 31, by BitWrangler

User metadata
Rank l33t++
Rank
l33t++

The early MX were "contenders" in that I think they got a couple of month leap on Intel for a "200" or "233" chip and the mags had them as performance leader on winstones, then the PII, MMX233 and K6 chips appeared and left them behind and Cyrix stayed "sort of close" for a year, leaving hope they'd manage a jump ahead again. But then their speed increases got further and further apart and everyone stopped hoping.

Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.

Reply 18 of 31, by Anonymous Coward

User metadata
Rank l33t
Rank
l33t

Did ST ever get access to the MII? I only seem to remember them doing up to the 6x86L. I think they got dropped after Cyrix was bought by NatSemi.

"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 19 of 31, by BitWrangler

User metadata
Rank l33t++
Rank
l33t++

I think I've seen MX with "Canada" markings on the backside which only makes sense if they were done at ST's Fab near Ottawa. I may be misremembering though.

Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.