| View previous topic :: View next topic |
| Author |
Message |
CHips

Joined: 01 May 2016 Posts: 834 Location: Switzerland
|
Posted: Tue Jul 14, 2020 7:14 am Post subject: QIC0 frequency |
|
|
Every resource says that QIC0 is 1.26 GHz, but I tested one and it looks like it is only a 1.2 GHz part...
Where does the 1.26 GHz info come?
Is the database wrong? Or does this processor need some special motherboard to be recognized as 1.26 GHz and work with 9.5 multiplier? _________________ Stelo.xyz Museum (CPU Collection and more) |
|
| Back to top |
|
 |
crazybubba64

Joined: 03 Jul 2018 Posts: 1371 Location: WI, USA
|
Posted: Tue Jul 14, 2020 7:57 am Post subject: |
|
|
Was the previous processor in the machine you ran it in a 1.2GHz part?
Sometimes Windows can get confused when you switch similar processors in a system.
Go to device manager and "uninstall" the processor. The computer will need to reboot, but it should re-detect the CPU correctly. _________________ My collection |
|
| Back to top |
|
 |
xsecret

Joined: 01 Feb 2004 Posts: 1846 Location: France
|
|
| Back to top |
|
 |
wren4777

Joined: 13 Dec 2016 Posts: 571 Location: Litija, Slovenia
|
Posted: Tue Jul 14, 2020 8:03 am Post subject: |
|
|
Is what windows 7 on a P3?  _________________ My WTB List
My collection
Have a nice day!  |
|
| Back to top |
|
 |
CHips

Joined: 01 May 2016 Posts: 834 Location: Switzerland
|
Posted: Tue Jul 14, 2020 8:15 am Post subject: |
|
|
| crazybubba64 wrote: | Was the previous processor in the machine you ran it in a 1.2GHz part?
Sometimes Windows can get confused when you switch similar processors in a system.
Go to device manager and "uninstall" the processor. The computer will need to reboot, but it should re-detect the CPU correctly. |
No, it was previously a Celeron 1333.
1.2 GHz is also shown in the BIOS.
| xsecret wrote: | | QIC0 is 1.26 GHz (9.5x133) Tualatin A4 |
Which source do you use to find out the identities from Sspec?
Could it be referenced as 1.26 GHz in some official Intel document, but only be 1.2 GHz on actual processors?
| linuxfanatic wrote: | Is what windows 7 on a P3?  |
Windows Vista Lol, though I am sure that Windows 7 would work fine as well. _________________ Stelo.xyz Museum (CPU Collection and more) |
|
| Back to top |
|
 |
xsecret

Joined: 01 Feb 2004 Posts: 1846 Location: France
|
Posted: Tue Jul 14, 2020 9:19 am Post subject: |
|
|
| CHips wrote: |
Which source do you use to find out the identities from Sspec?
Could it be referenced as 1.26 GHz in some official Intel document, but only be 1.2 GHz on actual processors?
|
Internal Intel DB. This specific Q-Spec is also directly linked to the commercial part number (RH80530GZ012512), so I'm pretty sure it's a 1.26 GHz part. In your case, you should probably try with another CPU with a half-multiplier. Maybe the motherboard don't support them. _________________ ES-Only Collector : http://www.engineering-sample.com
Universal Chip Analyzer (UCA) : https://x86.fr/uca / http://www.cpu-world.com/forum/viewtopic.php?t=34349 |
|
| Back to top |
|
 |
CHips

Joined: 01 May 2016 Posts: 834 Location: Switzerland
|
Posted: Tue Jul 14, 2020 9:36 am Post subject: |
|
|
I thought about that, however the ThinkPad T23 used to test it officially shipped with 1 and 1.13 GHz options, which have .5 multiplier. Now maybe for some reason only 9.5x would not work...
Also the CPUID Processor Number is 1200 MHz, isn't this the value "engraved" in the CPU? If it were actually a 1.26 GHz part, wouldn't it be IDed as 1266 MHz no matter whether the motherboard supports it or not?
Like if I put a Pentium 3 1400 in some 370 motherboard that somehow supports Tualatins but only at 100 MHz Fsb, it would still be IDed as 1400 MHz while actually running at 1050 MHz. _________________ Stelo.xyz Museum (CPU Collection and more) |
|
| Back to top |
|
 |
frag_
Joined: 17 Nov 2008 Posts: 4015 Location: Estonia
|
Posted: Tue Jul 14, 2020 9:51 am Post subject: |
|
|
Name is fused to CPU as processor brand string and can be read by CPUID leaves 80000002-80000004.
According to your cpuid dump it's clearly 1200.
Something must be wrong on Intel side +)
Or maybe it was relabeled? Can we see a photo? |
|
| Back to top |
|
 |
CHips

Joined: 01 May 2016 Posts: 834 Location: Switzerland
|
Posted: Tue Jul 14, 2020 10:13 am Post subject: |
|
|
Here are pictures. _________________ Stelo.xyz Museum (CPU Collection and more) |
|
| Back to top |
|
 |
xsecret

Joined: 01 Feb 2004 Posts: 1846 Location: France
|
Posted: Mon Jul 27, 2020 6:22 pm Post subject: |
|
|
PSU Name String (PNS) can be changed on ES but not on QS. On Intel DB, this QS is linked to a retail SKU that is listed on the official ARK (https://ark.intel.com/content/www/fr/fr/ark/products/27383/mobile-intel-pentium-iii-processor-m-1-26-ghz-512k-cache-133-mhz-fsb.html) but I have no trace of any PC actually using it. BTW, this CPU is listed with many SSPEC : SL6537 (A4) SL6AL (B1), SL6AA (B1). There is no less than 17 (!!) Q-Spec for that SKU, ranging from A3 to C0. They all come in pair for an unknown reason. Maybe some are tagged as 1.20 and other as 1.26 GHz. The QIC0 brother is QIC1.
I have checked the CPU-Z database and there is thousandths of Pentium III-M listed. I can confirm there is no 9.5x Pentium III-M. There is many 1.33 GHz and 1.20 GHz but no 1.26 GHz. On 1.20 GHz, the PNS also changed between A-Step and B-Step from "Intel(R) Pentium(R) III Mobile CPU 1200MHz" to "Mobile Intel(R) Pentium(R) III CPU - M 1200MHz".
So it seems Intel messed up something with the 1.26 GHz P III-M _________________ ES-Only Collector : http://www.engineering-sample.com
Universal Chip Analyzer (UCA) : https://x86.fr/uca / http://www.cpu-world.com/forum/viewtopic.php?t=34349 |
|
| Back to top |
|
 |
CHips

Joined: 01 May 2016 Posts: 834 Location: Switzerland
|
Posted: Mon Jul 27, 2020 7:13 pm Post subject: |
|
|
Very interesting! So no known working Pentium III-M 1266...
There is a SL6AA (1266/512) on Ebay, I am curious if this one is a true 1.26, or also only 1.2 like my QIC0. However $99.98 are asked...
Maybe something went wrong in the Intel factory and 1.26 GHz models ended up with 1.2 GHz dices...
Many 1.33 GHz though? Maybe that SL6CQ that I am chasing is actually common after all, but then I wonder where these folks found theirs... _________________ Stelo.xyz Museum (CPU Collection and more) |
|
| Back to top |
|
 |
CHips

Joined: 01 May 2016 Posts: 834 Location: Switzerland
|
Posted: Tue Sep 21, 2021 1:00 pm Post subject: |
|
|
Mystery solved, I never noticed a setting in the Bios (Power > Low/High Frequency), which allows to choose the max frequency.
For some reason the T23 does not choose 1266, even after a Bios reset due to dead batteries, but the option is available.
The setting changes the Processor Name, at least the frequency part, so it is not hard coded into the Cpu after all. It does not appear if using a production SL6A9 1200 MHz.
And there is no 1333 MHz option with a QOJ8, so unless we have another mystery (is QOJ8 really 1333?!), it looks like that a T23 is maxed out with a 1266 part, unless there is a way to Mod the Bios to fix this.
So here is the first ever 9.5x Pentium 3M Validation :p
http://valid.x86.fr/v5h6s7 _________________ Stelo.xyz Museum (CPU Collection and more) |
|
| Back to top |
|
 |
|