Jump to content
HWBOT Community Forums

The official HWBOT Team Cup 2015 - SC1: CPU Challenge thread.


Recommended Posts

  • Replies 228
  • Created
  • Last Reply

Top Posters In This Topic

  • Crew
I agree every submission ever made to hwbot should be deleted until the new cpu-z is out.

 

Sorry but if you dont know what we talk about you should better not comment.

I was referring to Romans post where it was about submissions with socket 7 systems and CPU-Z 1.72 and _not_ all subs ever made.

There is a bug somewhere introduced around 1.70 where CPU-Z doesnt detect half multipliers on that systems. There is a thread around here where things got confirmed and even debugged yet this new version hasnt released. Because of that its clear that current CPU-Z isnt a reliable verification method on that systems.

Link to comment
Share on other sites

I know exactly what you are talking about. How could I not, you've cried about it in 4 different threads now.

 

So just get on with it and stop whinging because you can't replicate what others are doing. There is no new CPU-Z yet, so there is no point to your constant complaining.

Edited by JunkDogg
Link to comment
Share on other sites

CPU CHALLENGE - SuperPI 1M 80486

2 questions:

1. "80486 family" means that the use of only the processor "intel" ((AMD, Cyrix have names Am486 Cx486), that is the logic of "486 series" but not "80486 series"

2 If it is not only intel - is allowed 5x86 processor? (this the fourth generation processors too)

 

The list of processors can be found here: http://hwbot.org/hardware/processors#key=80486

Link to comment
Share on other sites

  • Crew
I know exactly what you are talking about. How could I not, you've cried about it in 4 different threads now.

 

CPU-z is our verification system that we trust. It's hard to believe when it's not working the way that it should especially on systems which are that old and worked for years without problems. Now we have the situation that verification isnt working correctly. And thus all verifications made with those versions can be considered worthless because you cant trust them. Sorry but it's simply unfair for future benchers who dont profit from software bugs. It's not the way that I want to sound like a teacher. My english abilitys are bit limited so I really dont want to write essays and explain everything in detail. But I still want to say my opinion. It's never meant personaly. So it's not against you or classicplatforms in general. But just go ahead and keep attacking me.

Link to comment
Share on other sites

CPU-z is our verification system that we trust. It's hard to believe when it's not working the way that it should especially on systems which are that old and worked for years without problems. Now we have the situation that verification isnt working correctly. And thus all verifications made with those versions can be considered worthless because you cant trust them. Sorry but it's simply unfair for future benchers who dont profit from software bugs. It's not the way that I want to sound like a teacher. My english abilitys are bit limited so I really dont want to write essays and explain everything in detail. But I still want to say my opinion. It's never meant personaly. So it's not against you or classicplatforms in general. But just go ahead and keep attacking me.

 

This is possibly the most confusing post I've ever read.

 

Please do not now sit there and pretend to be the victim, you came out and pretty much called everyone using the current version of CPU-Z cheats. Now you are back pedaling.

 

There is no current fix for this(if it really is a bug,as it can't be proven until new version is made) so until then there is really no point to complaining about it mate.

 

Once the new version is here and the bug can be verified, it's a few peoples word against other people's word.

 

I honestly would be suprised if we see it before the end of TC, then what delete every submission in the comp that has used the old version? There will be anarchy.

Link to comment
Share on other sites

JunkDogg

On one side you are right.

But on the other... let me escape to urgently look for the motherboard (and the CPU) on which CPU-Z shows a 1.0 instead of 1.5 until output new version ;)(joke)

How better to do I do not know, but obviously impossible results better do not accept.

Edited by RomanLV
Link to comment
Share on other sites

CPU-z is our verification system that we trust. It's hard to believe when it's not working the way that it should especially on systems which are that old and worked for years without problems. Now we have the situation that verification isnt working correctly. And thus all verifications made with those versions can be considered worthless because you cant trust them. Sorry but it's simply unfair for future benchers who dont profit from software bugs. It's not the way that I want to sound like a teacher. My english abilitys are bit limited so I really dont want to write essays and explain everything in detail. But I still want to say my opinion. It's never meant personaly. So it's not against you or classicplatforms in general. But just go ahead and keep attacking me.

 

Uh uh uh, Classicplatforms hasn't said boo in this thread. Junkdogg's opinion is his own and may or may not reflect others opinions on the team. Keep it between you and him please. Tia:)

Link to comment
Share on other sites

I might found a solution to the issue that new cpu-z version is not sensing correct the half multipliers. Excuse me in case reported again earlier.

 

If we go to the cpu-z folder and edit the cpu-z configuration file by set the below values to "0" , then the frequency displayed by cpu-z is the correct one.

 

DMI=0

DISPLAY=0

BUSCLOCK=0

SPD=0

 

All the rest untouched.

 

checked from cpu-z 1.69.2 version until 1.71.1 and works.

Submission also works fine.

Is that ok submitting results this way?

 

I see that Remarc & LuDec already know this and submitted their results.

Edited by varachio
Link to comment
Share on other sites

  • Crew

This is possibly the most confusing post I've ever read.

 

Well this is because _Im_ a very confused person. ;)

 

Please do not now sit there and pretend to be the victim, you came out and pretty much called everyone using the current version of CPU-Z cheats. Now you are back pedaling.

 

I never said they cheated, I just made them aware of a bug. Its disappointing that they didnt start to investigate what could be wrong. Now we see that those 50Mhz submissions got deleted because _they were_ faulty.

 

There is no current fix for this(if it really is a bug,as it can't be proven until new version is made) so until then there is really no point to complaining about it mate.

 

This bug has been confirmed and already debugged. I posted in every of my "complains" the link to the bug thread. So to make it even more clear and because I still think that you actually have no clue what we talk about:

 

1. On motherboard or in the manual there are all possible multiplier settings listed. Except there is no 1.0 multi, so you set the lowest which is 1.5.

2. motherboard starts up showing 75Mhz in post(1.5x50Mhz)

3. CPU-Z starts up saying 1.0x50Mhz, strange...

 

So you dont get curious? You dont check tools like WCPUID or CPUCool?

Ok at this point, we still can say those things can happen but after people commented under your score stating that there is probably a bug you still dont react?

 

I really hoped that new CPU-Z is out now because of Windows 10 and Skylake but it seems something hold it back.

Link to comment
Share on other sites

  • Crew
Thank

 

By the way, there is an incorrect database. Socket3 are not only AMD. Moreover (I posted above) - names 80*** have just intel. AMD have names Am*** , Cyrix - Cx***

Administration - can fix it in the database?

 

We are about to get everything done. There are just the Cyrix 486 Cpus missing.

 

Massman still needs to update the restrictions to allow Cyrix, Intel and UMC ones.

Link to comment
Share on other sites

1. On motherboard or in the manual there are all possible multiplier settings listed. Except there is no 1.0 multi, so you set the lowest which is 1.5.

2. motherboard starts up showing 75Mhz in post(1.5x50Mhz)

3. CPU-Z starts up saying 1.0x50Mhz, strange...

 

Well, not everything is written in the mainboard guide, you know? I managed to change the FSB to 7.14MHz and therefore I get 10.7MHz on friend Asus mobo:

http://valid.canardpc.com/show_oc.php?id=348684

http://img63.imageshack.us/img63/2005/asustxp4x22sr5.jpg

 

Beat that... or even come close for that matter :) And how I did it? Just reading wrong, what is written into the manual :)

 

(hint - first I flashed bios to mobo, that it does not belong to, then I set undocumented variant of switches and - voila! When checking for 5min in the manual, what went wrong, the mobo got the time to post :) Yes, it took 5min to even show the first letters of postscreen at 10.7MHz... :) Another problem - not all windows can boot at 10.7MHz... And another problem - PS/2 keyboard and mouse did not worked :)))) So, you see... sometimes you have to overcome quite a bit of problems to achieve something... And I know now, how to get a x1 multiplier, witch translate into 7.14MHz CPU clock. That will be something! ;) )

Link to comment
Share on other sites

  • Crew

So to achieve this you basically have to trick your pll thinking it has to change its FSB outputs to ISA outputs, as there are no other components in the system which the pll delivers with such a low bus clock. But that has to happen internal as the connection between CPU and PLL doesnt change. While I dont know if the pll can internal reroute its clock signals, it would be interesting to know if changing jumpers to your found undocumented setting also set your bus to 7Mhz with original bios!

If so you really found something cool! If not, its more likely that the other Bios transformed your system to a complete mess. Means all sources where CPU-Z detects your clock frequency are now basically upside down.

I experienced something similar when changing default clock generator (14,318Mhz) to something faster. I checked bus frequency with a DMM and its improved correctly but CPU-Z didnt show the same results as my DMM. So why? Its actually easy. I forgot that system time also relies on that 14,318Mhz. So time got "improved" as well and was running freaky fast.

Link to comment
Share on other sites

Well, I was asked my friend, for who I recapped the board to use as homemade server back then, if I could have and I he agreed, so in about two to there weeks I should have it.

IIRC he said he tried to duplicate the results, but failed. So I quess that the wrong bios is necessary to achieve that feat.

 

And as you can see, CPU-Z fail to show any clock at all, claiming the FSB is zero... :) Lowering the clock crystal will be the next step to get lower that 7.14 :D

Link to comment
Share on other sites

QuickFast

QuickFast says:

 

Massman said: The challenge is to find the CPUs that both validate and clock really low :)

 

 

 

So is part of the challenge to get a valid submission?

 

 

 

 

 

I'll make this more clear, in the CPU low clock many of the subs are rejected by CPUz if you follow the links. Are these still legitimate

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...