3DFX Voodoo5 5500 & ASRock 4CoreDual-Sata2 New Experiment!

Aperto da ciacara, 03 Aprile 2011, 17:03:27

Discussione precedente - Discussione successiva

0 Utenti e 1 Visitatore stanno visualizzando questa discussione.

ciacara

Un'alta domanda giusta ;)

Come ho scritto all'inizio del thread:
 
Citazione"Dopo ripetuti fallimenti son riuscito a trovare il giusto equilibrio e le giuste modifiche Hardware per installare e rendere funzionale una Voodoo 5 5500 normale su un' ASRock 4CoreDual-Sata2 con un Q6600"
[/B]

Nell'esperimento precedente non avevo apportato modifiche al VAgp, quindi, la tensione che mi serviva per far partire la Voodoo con agp@4x dovevo ottenerla con il potenziometro.
Solo dopo aver avviato il sistema (Abit AN7) e caricato il sistema operativo, potevo aumentare il voltaggio a 3.3v "A CALDO" per dare la giusta alimentazione alla Voodoo e rendere il tutto stabile...cosa che, con un voltaggio inferiore  non potevo ottenere.

In questo Esperimento, ho apportato delle Vmod direttamente alla Scheda Madre (Asrock), quindi, non ho avuto bisogno di un potenziometro per dare la giusta tensione all'AGP e farlo partire @4x e far riconoscere la Voodoo, mi è bastato aumentare il voltaggio dell'agp per ottenere lo stesso risultato dell' Abit AN7.


La differenza sostanziale tra i due esperimenti  è nella reazione della Scheda Madre ad una Alimentazione Esterna/Forzata dei 3.3v:

L' Abit AN7, forse per un sistema di protezione sulle tensioni, accettava questa sovralimentazione esterna come se fosse erogata direttamente dalla Scheda Madre, quindi, non richiedeva  ulteriori VMod.

L' Asrock 4CoreDual-Sata2 reagisce diversamente a questa sovralimentazione esterna/forzata.
Si creano degli scontri di tensioni che danno vari tipi di problemi, che vanno dalla corruzione dei dati al malfunzionamento dei vari componenti ed un surriscaldamento dei Mosfet addetti a gestire queste tensioni.

Per questo motivo ho parlato di "giusto equilibrio", proprio perché ho dovuto bilanciare il tutto per renderlo stabile e funzionale.

Il trimmer che ho aggiunto al Riser Agp mi è servito a trovare questo equilibrio, anche se l'influenza che ha sui 3.3v è minima, il sistema, almeno nel mio caso, non funzionerebbe a dovere e risulterebbe instabile senza questo componente che ho aggiunto.
 

Spero di aver dato una giusta risposta alla tua/vostra domanda.

Aggiunto al Thread:

1.Test Benchmark 3Dmark 2001
2.Risposte/Chiarimenti/Spiegazioni date a Radu
3.Fornitore Riser Universale Agp


Resto online ancora per un pò e poi vò via. ;)




High right question ;)

As I wrote at the 'beginning of the thread:
 
Citazione"After repeated failures, I managed to find the right balance and the right hardware changes to install and make functional a Voodoo 5 5500 on a normal 'ASRock 4CoreDual-Sata2 with a Q6600"
[/B]

In the previous experiment I had made changes to VAgp, therefore, the tension that I needed to start with the Voodoo @ agp 4x I get with the potentiometer.

Only after starting the system (Abit AN7) and loaded the operating system, could increase the voltage to 3.3v "HOT" to give the right power to make it stable and Voodoo ... thing, with a lower voltage could not get.

In this experiment, I made some Vmod directly to the motherboard (Asrock), so I did not need a pot to give the correct voltage and AGP 4x @ to start it and to recognize the Voodoo, I simply increase the voltage of 'agp to get the same result of' Abit AN7.

The main difference between the two experiments is in reaction to an External Power Motherboard / Forced of 3.3v:
The 'Abit AN7, perhaps for a security system on the tensions, accept this as if it were external boost delivered directly from the motherboard, therefore, did not require further vmode.

Asrock 4CoreDual-Sata2 reacts differently to this external boost / forced.
This creates tensions of clashes that give various kinds of problems, ranging from data corruption to the failure of the various components and overheating of the MOSFET to manage these tensions.

For this reason I talked about "balance," because I had to balance everything to make it stable and functional.

The trimmer I added to the AGP Riser helped me to find this balance, although the influence it has on the 3.3v is minimal, the system, at least in my case, would not work properly and would be unstable without this I added .
 

I hope I have given a correct answer to your question. :)

Added to the thread:

1. Test Benchmark 3Dmark 2001
2. Responses / Clarifications / Explanations given to Radu
3. Supplier Universal AGP Riser
[/COLOR]

radu14m

the score in 3dmark2001 is low...it's look like a voodoo4 score.
You can try a voodoo4 and see what is the score that you get !
 
Are you sure that both VSA's are working correct ?:beerbeer:
 
can you measure the voltages of the VSA's ?
( should be 2.8Volts )
:h

ciacara

...
#32
Citazione di: radu14m;175561the score in 3dmark2001 is low...it's look like a voodoo4 score.
You can try a voodoo4 and see what is the score that you get !
 
Are you sure that both VSA's are working correct ?:beerbeer:
 
can you measure the voltages of the VSA's ?
( should be 2.8Volts )
:h

Hello Radu,
Are you sure that you are not confusing with 3DMark 2000? :beerbeer:
 
Read this thread:
 
https://www.3dfxzone.it/enboard/topic.asp?TOPIC_ID=401&whichpage=3
 
Also read this: ;) :)
 
https://www.3dfxzone.it/enboard/topic.asp?TOPIC_ID=6626&whichpage=1
 
 
Again congratulations for the work you have done:app: ... now you should run for the Power Vmod direct 3.3v ;)... now you know how to do this. :)
 
However, I have made ​​tests with V-sync on, and Windows 7 ...I will try to do some testing with Windows XP and maybe with the Windows 98se.
 
 
The 3DMark 2000, which is more suitable in my opinion, things change ...;)
 
 
However the system works well and no blocks, just look at the VideoTest I did ... all the videos you've watched?
 
 
Do you have any example to show me please? ... thanks :)
 
 
Ora mi tornano in mente anche le parole di Glide...;)
 
E Glide, cosa ne pensi?

ciacara

Piccola revisione al Primo Post.
 
Ho separato la parte in Italiano da quella Inglese.
 
Adesso c'é una lettura scorrevole della Guida. ;)
 
 
 
Small revision to the First Post.
 
 
I separated the English from the Italian part.
 
 
Now there is a smooth reading of the Guide. ;)
 
 
Buona Serata, Ciao.

radu14m

No, i speak about 3dmark2001.
look here:
http://www.falconfly.de/cgi-bin/yabb2/YaBB.pl?num=1308423503/5#5
 
score is 2800-2900 points, with a single core CPU.
You are using a quad core, so there should be more inside :)
 
Think about your projekt: voodoo5 and Abit NF7S :) - you get there with a single core CPU over 3000 points...
 
Maybe the voodoo is running at 33MHz ?  Like a PCI version ?
I have the same situation, with a v5 PCI and a quad core phenom, i get around 2500 points....
 
Another point: is the transfer rate AGP2x or 4x ? ( on your projekt )
Can you check this with a program ?
 
AFAIK there is a hardware bit that have to be enable via a resistor to set the AGP4x transfer rate....check please the VSA datasheet
 
Thanks ciacara...
 :merlin:

ciacara

#35
in this post, you compared a voodoo 5500 to a score of  voodoo of 4500 ...

Citazione di: radu14m;175561the score in 3dmark2001 is low...it's look like a voodoo4 score.
You can try a voodoo4 and see what is the score that you get !


in this post, you compared 5500 with a score of other voodoo 5500, including the PCI 5500 is your ...


Citazione di: radu14m;175631No, i speak about 3dmark2001.
look here:
http://www.falconfly.de/cgi-bin/yabb2/YaBB.pl?num=1308423503/5#5
 
score is 2800-2900 points, with a single core CPU.
You are using a quad core, so there should be more inside :)
 
Think about your projekt: voodoo5 and Abit
NF7S
:) - you get there with a single core CPU over 3000 points...

Therefore, my it is not a score Voodoo 4500.



This is the manual of the AN7 'experiment 1 (not NF7S):

http://file.abit.com.tw/pub/download/manual/english/an7.zip

"(10) ATTENTION: This motherboard does not support 3.3v Agp card. Use only 1.5 or 0.8v Agp Card."

This is the manual of the 4CoreDual-Sata2 experiment 2:

http://europe.asrock.com/downloadsite/manual/4CoreDual-SATA2.pdf

"2.6 - Please do NOT use a 3.3V AGP card on the AGP slot of this motherboard!
It may cause permanent damage! For the voltage information of your
AGP card, please check with the AGP card vendors."


No motherboard was compatible with AGP2X ... But they work.

No Voodoo 5500 was compatible with AGP@4x ... But they work.

CitazioneAnother point: is the transfer rate AGP2x or 4x ? ( on your projekt )
Can you check this with a program ?

Any utylity I used (Everest, CPUZ, v-control...) detects the AGP 4x enabled, both the mother and the Voodoo card 5500.
I posted everything I've done.

I did these experiments to run a motherboard is not compatible with a AGP2X voodoo5500 AGP4X not compatible with ... and both work.

I made these experiment not to use 3DMarck 2001 and to 5000 points.

CitazioneAFAIK there is a hardware bit that have to be enable via a resistor to set the AGP4x transfer rate....check please the VSA datasheet

The trimmer of the Universal AGP Riser NOT ACTIVE AGP@4X on the Voodoo.

This serves to lower the voltage of 3.3V, even if just to avoid, in my experiment, a system instability.

The potentiometer of 50k AN7 experiment, active AGP@4X the motherboard and starts the Voddo AGP4X mode.
Even here I posted everything I've done.

The trimmer of the 4CoreDual-Sata2 VAgpmod serves to provide power to voodoo to turn on and boot mode AGP4X.
Even here I posted everything I've done.

According to various guides, manuals, documents, datasheets that are online, this IS NOT POSSIBLE! ... Yet, everything works perfectly.

A friend told me:

Ciacara, if not the things you do not understand.

Now the question is, what's the problem? ...

Now Radu, it's time that you burn some voodoo, cpu, motherboard to understand. ;) :)[/COLOR]


Qualsiasi utylity che ho usato, everest, cpuz, v-control rileva l' AGP @4x attivato, sia per la Scheda madre che per la Voodoo 5500. Ho postato tutto quello che ho fatto.

Ho fatto queste Modifiche non per usare 3DMarck 2001 e fare 5000 punti.

Ho fatto questi esperimenti per far funzionare una scheda madre non compatibile con agp2x con una una voodoo5500 non compatibile con agp4x... e funzionano entrambe.

Il Trimmer del Riser Agp Universale NON ATTIVA L'AGP4X DELLA VOODOO.

Questo serve per abbassare il voltaggio dei 3.3v, anche se di poco, per evitare, nel mio esperimento, un instabilità del sistema.

Il Potenziometro da 50k del AN7 attiva l'Agp4x della scheda madre e fa partire la Voddo in modalità Agp4x.
Anche quì Ho postato tutto quello che ho fatto.

Il Trimmer della VAgpmod del 4CoreDual-Sata2 serve per dare i Volt necessari alla voodoo per avviarsi in uno slot Agp@4X in modalità 4x.
E anche quì Ho postato tutto quello che ho fatto.

Secondo le varie guide, i manuali, documenti,  datasheet che ci sono online, tutto questo NON  E POSSIBILE!... eppure, funziona tutto perfettamente.

Un amico mi ha dettio:

Ciacara, le cose se non le provi non le capisci.

La domanda è: qual è il problema? ...

Radu è ora che anche tu bruci qualche voodoo, cpu, scheda madre per capire. ;) :)


Buona serata. :)


P.S. Aggiunti al primo post Preview Video. :okduo:

Grazie Glide! :risp:

radu14m

Yes , now it's time for me to burn and destroy ( hopefully not ) some hardware...
 
I'm looking now for a suitable Motherboard to start my experiments too...you give me hope now with your results :)
Thanks ciacara for this experiments :)
As i say before: A lot of people are talking in the 3dfx forums, about differents projekts....but when it"s come to the "doing" moment...very very few are up to the task are are doing something real.
Thanks that you are one of the few who are puting the dreams into the reality...
 
:)
Thanks again for the gorgeous experiment :beerbeer:
Radu.

ciacara

Thanks again Radu. ;)

The advice I can give is to use VGA card AGP2X old, not to test directly with Voodoo.

I'll tell you why I unfortunately corpses of 3dfx, CPU, motherboards in the closet ...:wall:

Proof that you can do:

1. increase VAgp v.1, 88 / 1.87
2. install a heatsink on the NB and MOSFET AGP
3. test the change without the direct power of v.3, 30 on the AGP

If all goes well then you can go to the next steps ...

I wish you good luck :kazim: ;) , and do not say bye because you have to go back to this thread to celebrate  your success!:beer: ... or to ask for advice to the Master Modder 3DFX or a blessing for your project! :risp: :D

:hello:

ciacara

#38
Corretta immagine Riser AGP Universale

Aggiunto Riva Tuner Test ai Benchmark
:)

Messaggio dalla redazione:
Contenuto non disponibile in quanto rimosso da server esterno o server esterno off line

radu14m

Thanks Master Modder 3DFX  ;)
 
I will come back and post my results.
Be sure, there will be question from my side when i will start the projekt...
I will come here for advices :circle:
 
Radu