Tbred step B voltage tolerances??

[designgirl]

i've tried looking at the technical documents to find this, but not very good at reading this type of stuff.

what's the maximum voltage that these chips can take ( within spec )?? The reason is, I want to overclock but only as far as I can within the voltage tolerance that AMD has set.

thanks

 

[MrC4]

I do not know if you will find any official voltage tolerance from AMD other than the spec of 1.65v. Personally, if you have good air cooling, you should be able to increase your core voltage up to 1.85 without any trouble.

Good luck!

 

[pillage2001]

Anything above what the factory default is out of spec. I would personally not go above 1.8v for the chip. 0.13micron is too delicate.

 

[chizow]

Well...the Tech sheets do tell you the maximum temps the chips can handle....85C for "V" T-breds (2200+ and higher)....90C for "T" T-breds (2100+ and lower).

Chiz

 

[bunnyfubbles]

Originally posted by: pillage2001
Anything above what the factory default is out of spec. I would personally not go above 1.8v for the chip. 0.13micron is too delicate.

I've seen several instances where 2v was used without blinking twice. I think these Tbred Bs are pretty tough for .13...

 

[pillage2001]

Originally posted by: bunnyfubbles

Originally posted by: pillage2001
Anything above what the factory default is out of spec. I would personally not go above 1.8v for the chip. 0.13micron is too delicate.

I've seen several instances where 2v was used without blinking twice. I think these Tbred Bs are pretty tough for .13...

Let it run for a few months before it breaks down. There's a term for it but I've forgotten what it's called.

 

[MrC4]

Let it run for a few months before it breaks down. There's a term for it but I've forgotten what it's called

Spontaneous Combustion??

 

[pillage2001]

Originally posted by: MrC4

Let it run for a few months before it breaks down. There's a term for it but I've forgotten what it's called

Spontaneous Combustion??

:Q.......definitely not that. It takes just a few sec to do that.

 

[ChampionAtTufshop]

Originally posted by: pillage2001

Originally posted by: MrC4

Let it run for a few months before it breaks down. There's a term for it but I've forgotten what it's called

Spontaneous Combustion??

:Q.......definitely not that. It takes just a few sec to do that.

gate oxide breakdown?

 

[nemo160]

electromigration perhaps?

 

[BD231]

If you put to much voltage into a P4, you pretty much burn up "copper innterconnects" as far as I can remember, you also kill transistors by hampering their ability to turn on and off correctly. The Gate Oxide phrase rings a bell as well, can't remember what it means though. In any case this eventually leads to death of the CPU at any voltage and clockspeed. Why this would be any different for the T-Bred is beyond me, especially when you consider the CPU is working harder clock for clock than a P4. Possibly because the pathways are much shorter on the T-Bred?, either way I would not go past 1.70v myself, this is not a Palimino were talking about here.

 

[AnImuS]

you guys are making me scared of running my cpu at 1.9v

 

[Ketchup]

It was a while before we heard about electomigration killing the P4. The t-bred's haven't really been out that long, so it will be a while before we find out how sensitive these chips are to voltage.

 

[chizow]

Originally posted by: ketchup79
It was a while before we heard about electomigration killing the P4. The t-bred's haven't really been out that long, so it will be a while before we find out how sensitive these chips are to voltage.

Its ok, if I kill mine I'll just pick another one up with my pocket change

Chiz

 

[MichaelD]

Originally posted by: nemo160
electromigration perhaps?

Ding. Give the man a cigar. If you run at 1.850 v or above, give your chip 6-9 months to live, if that. Most of us upgrade more often than that anyway, though.

Personally, I overclock, but I DON'T like running things on the edge of stability/breakdown. It's nice to see "just how high I can get this mutha to go" for a day or so...just to see, but heck, I want my $100-worth (or whatever I paid for XXX piece of equipment.

For sake of arguement, you have a B TBred 2100 and good air cooling. You can get 2.3-2.4 GHz out of it, but only at 1.8-1.85 volts..temps are high and the system is about 85% stable. Well, crank it down, just a notch to 2.1-2.1GHz at a lot less voltage, like 1.70 at not only does the chip run a lot cooler, but the thing will probably last a good long time. That's my line of thinking and what I'm currently doing with my B TBred 2100.

 

[Shimmishim]

Originally posted by: chizow

Originally posted by: ketchup79
It was a while before we heard about electomigration killing the P4. The t-bred's haven't really been out that long, so it will be a while before we find out how sensitive these chips are to voltage.

Its ok, if I kill mine I'll just pick another one up with my pocket change

Chiz

blah!

 

[pillage2001]

Originally posted by: nemo160
electromigration perhaps?

I think this is it.

 

[Parasitic]

I think 1.850V should probably be the upper limit anyhow.

 

[Shimmishim]

well... we'll see.. i'm at 1.875V on this baby...

 

[ProviaFan]

I'm just going to bump this thread with my question instead of starting a new one; hopefully someone will feel benevolent and answer.

At this time, my Athlon XP 1700+ Thoroughbred B processor is running at 159FSB * 12.5 multiplier, for just a few MHz under 2000; the core voltage is 1.75. It's been running Prime95 for a few hours now without reporting any errors, but I plan to run Prime95 again overnight to see if it remains stable. Under this load, the temperature reaches 46°C on average, while when running at default settings, the temperature is around 38°C (I'm using a lapped Thermaltake Volcano 7 with Arctic Silver 3). I'm afraid with these voltages and temperatures, that my processor might be in danger. Am I just needlessly worried, or should I back off and run at 1800MHz at default voltage and lower temperatures?

 

[ChampionAtTufshop]

you are fine i believe
temps look ok and voltage isnt that high

 

[pelikan]

46C is very low, no problem at all. 1.75V should also be fine. The chips I've read about that died were at 1.95V and over.

 

[NesuD]

Originally posted by: MichaelD

Originally posted by: nemo160
electromigration perhaps?

Ding. Give the man a cigar. If you run at 1.850 v or above, give your chip 6-9 months to live, if that. Most of us upgrade more often than that anyway, though.

Personally, I overclock, but I DON'T like running things on the edge of stability/breakdown. It's nice to see "just how high I can get this mutha to go" for a day or so...just to see, but heck, I want my $100-worth (or whatever I paid for XXX piece of equipment.

For sake of arguement, you have a B TBred 2100 and good air cooling. You can get 2.3-2.4 GHz out of it, but only at 1.8-1.85 volts..temps are high and the system is about 85% stable. Well, crank it down, just a notch to 2.1-2.1GHz at a lot less voltage, like 1.70 at not only does the chip run a lot cooler, but the thing will probably last a good long time. That's my line of thinking and what I'm currently doing with my B TBred 2100.

That is exactly what i have done with mine as well.

 

[eno]

1800+ TB B @ 208fsb/9.5 and 1.85v full load 38c with Slk800/SmartCase2 and ArticSilver3

 

[flyboyait]

Originally posted by: MichaelD


Ding. Give the man a cigar. If you run at 1.850 v or above, give your chip 6-9 months to live, if that. Most of us upgrade more often than that anyway, though.

Has the TBred been out 6-9 months?? any links to these things dying over 1.85? im running mine at 1.888-1.9 and uh... yeah i dont really wanna fry it and 2220mhz feels considerably faster than 2000.... any articles or anything that talk about this limit?

-FBA