XS HWBot Team

[Don_Dan]

I heard some guys will be benching for the Team Cup...
I'll send you a PM later. And don't worry about K7 32M, someone else is working on it.

[l0ud_sil3nc3]

count me in for something, not sure what atm, but I will fill in what is left.

[BeepBeep2]

backups backups backups backups
I should have LN2 by the last week of June since I finally got a job.

I'm weaker in K8 than I would have liked...

[Don_Dan]

Very nice scores guys! And awesome Computex action, Solid Snake!

And thanks a lot to BeepBeep2 for the first XS submission to the Team Cup!

[BeepBeep2]

Okay, so that was horrible. I wasted about 7 Lbs of DICE to accomplish nothing.
I was able to pull pot temp down to -40c and still boot, but using clockgen to change clocks meant a lockup? O.o Then, CPU temp readings were around 30c in software...

At -20c, I had about 10c idle in software but it was no better than what I had with straight ice in a bucket...
Then I was unable to pull the same clocks I could at that temp than I could with ice in a bucket...

I think I'm just gonna move on to K10 and revisit K8 again with LN2.

[BeepBeep2]

Well, added K10 32M on DICE entry to the Team Cup SC4.
Currently 2nd, I expect 1st with LN2 I might get creative and put LN2 on dimms too, but I don't have a ram pot...might just pour LN2 directly on them, or in aluminum foil over them

[ZenEffect]

what did you use as a buffer? dice should be pulling down to -65 to -75c if done right.

[BeepBeep2]

what did you use as a buffer? dice should be pulling down to -65 to -75c if done right.

When I said -40c, I was talking about K8, my A64 X2 5600+ Windsor, which most have a 0 to -15 cold bug...I was able to pull pot base to -40 and still boot up, but I was unable to change frequency for example in windows or run 32M.

I use Klean Strip S-L-X Denatured alcohol, just enough to fill holes in the pot and a little more. I've used it since doing dice for the first time. Doesn't get thick or anything, very thin even when there is a lot of DICE in the pot. Bubbles like crazy...essentially it is 1/2 ethyl alcohol, 1/2 methanol...freeze point about -100.

For K10, average base temps were about -60 idle and -53 load, -62 idle when going for CPU-Z valid with 1 core activated. Not sure how accurate my probes are, but they are accurate at room temperature. I had mostly a powder of dice, but some were small chunks.

Checking back on my 1100T DICE result on HWBOT I reported -59 idle and -53 load. My probes read about -169 to -173 full pot idle on LN2.

[ZenEffect]

what does your probe read when you dip it into the ln2? I only have 1 probe that reads below -190

[BeepBeep2]

I believe it reads about -180, will check again on Monday.
Bought badatgames18's 30L dewar, late 1950s model from Linde/Union Carbide. No dings in it, but shipping was rough, hope the vacuum is still good.

If things go as planned, then I'm looking at 955 BE on LN2 on Monday

[BeepBeep2]

I take it everyone is too busy to bench...not many new submissions for anything at all on the bot.
I'm taking on SC4 alone and nobody is overclocking for SC2 yet ><

[I.nfraR.ed]

I know I'm not from the hwbot xs team, but writing this just for the heads up.

Well, my team is very small (4 people) and I'm going to handle all AMD's, but have no intel chips (except s1155) and probably my team will have 2 submissions in SC2 or none at all. We just don't have many Core2 cpus if any at all.
So, just bench for fun .

[BeepBeep2]

I know I'm not from the hwbot xs team, but writing this just for the heads up.

Well, my team is very small (4 people) and I'm going to handle all AMD's, but have no intel chips (except s1155) and probably my team will have 2 submissions in SC2 or none at all. We just don't have many Core2 cpus if any at all.
So, just bench for fun .

You've done some really impressive stuff with the AMD chips

I don't have K6, I have a low-end 1600+, but to overclock it I'll have to use a pinmod.

[BeepBeep2]

955 Black Edition with CBB at -110, CB at -120.

HT Link @ 1000 helps, but every time I screw up I have to heat cycle the pot again, wasting a lot of LN2 and I haven't even gotten to Windows yet.

[Don_Dan]

955 Black Edition with CBB at -110, CB at -120.

HT Link @ 1000 helps, but every time I screw up I have to heat cycle the pot again, wasting a lot of LN2 and I haven't even gotten to Windows yet.

Well, that sucks.

That doesn't sound like a Phenom II at all, maybe it's a Core 2 in disguise...

[BeepBeep2]

I.nfraR.ed kind of tipped me to something grand master chew* said once about a cracked core and CPUs coldbugging before dying completely.

On try 1, I heard some weird cracking noises when pulling temps down...Of course, I wondered what it was but otherwise payed no attention to it. Day 2, same coldbug but I couldn't get it to post again after warming a second time.

Hasn't posted since...threw in 565 BE and it booted right up.

My parents aren't very happy and I'm out of money, but hopefully I can get one more of these before the end of June.

Chip died completely at stock settings under LN2...

[Don_Dan]

I'm sorry to hear that, any chance of RMAing that chip?

I found another kit of BH-5. It's only 2x256MB unfortunately but it might still be useful for something...

[BeepBeep2]

Nah, I'm not gonna RMA it. After 1.8v on dry ice I don't think it would be fair. Taking the IHS off it instead now.
I have another 955 and a 965 coming so not all hope is lost. They are so cheap these days >< not disposable but very affordable.

EDIT:
Well that was horrible.
A little of the die came out okay, the rest got stuck to the IHS and it sounded exactly like it did when I heard the popping on LN2 as I pried it. It even ended up pulling part of the PCB. ><

I heated it to 200c, I guess that wasn't enough to melt the solder soft enough. It was pliable, kind of sticky.

[G.Foyle]

IHS solder is most certainly indalloy (indium and tin) and melts at around 120C - maybe the die was cracked already, or maybe temp was too high and solder between die and substrate melted?

[Hondacity]

mass production do not use pure indium as it melts at a higher temp...

though they use indium alloys, I think they still use epoxy to make sure it sticks onto the die. thats why prying off the ihs is incredibly difficult.

[BeepBeep2]

I don't know...IHS temp was around 150c when I tried.

I couldn't pry it off easily. PCB bent a lot.

I had 3 of the blades that go into box cutters/utility knives whatever you call them and I wedged 2 under the IHS on each side, set oven to ~200c and waited until CPU was at 180 before I tried.

It took like 5 tries and it turned out horribly...

[BeepBeep2]

Alrighty everyone...

I'm getting a fill on Monday, when my two Phenom II X4's are supposed to be here.

If anyone has decent DDR in North America, that can ship in 3 days or less before the end of the month for the sake of the team then I will OC my T-Bred with a pinmod.

If not then I will just run to get a submission in, in which case we will be LAST (~5 pts vs up to 25) as I have generic DDR and the slowest CPU in the K7 lineups.

Birthday on the 3rd, would like to celebrate with an XS SC4 top 3.

[Don_Dan]

Nah, I'm not gonna RMA it. After 1.8v on dry ice I don't think it would be fair. Taking the IHS off it instead now.
I have another 955 and a 965 coming so not all hope is lost. They are so cheap these days >< not disposable but very affordable.

I used to think like that as well, but after seeing how many of the top overclockers just send their chips off to RMA, I changed my mind.
Not that it matters anyway, all my hardware is old and not covered by any warranty at all.

Alrighty everyone...

I'm getting a fill on Monday, when my two Phenom II X4's are supposed to be here.

If anyone has decent DDR in North America, that can ship in 3 days or less before the end of the month for the sake of the team then I will OC my T-Bred with a pinmod.

If not then I will just run to get a submission in, in which case we will be LAST (~5 pts vs up to 25) as I have generic DDR and the slowest CPU in the K7 lineups.

Birthday on the 3rd, would like to celebrate with an XS SC4 top 3.

I still hope there will be a K7 submission, but I sent you a PM concerning the DDR memory.

[BeepBeep2]

Thanks.

[BeepBeep2]

...aaaannndd I am over the 300 point hump. 400 next