Printable View
Sorry, I missed something there, I thought you said your current room temp was 24-25 and the current cpu temp 28? :lol:
Anyway, it's impressive and I wish I could afford something like it.
But at least I know what kind of stuff I'll be getting next time I've got a lot of money to spend. Wahey.
That was 28C whilst running at 100% CPU load.Quote:
Originally Posted by SnnY
At idle, she's around 22C which would be below my room ambient (i.e., not possible).
Still, I'm a happy boy.
D'oh
And I'm slow :lol:
Right, then.
I've been making attempts to discover the upper limits of my OC.
Amazingly (amazing because I have no idea about the mechanics of the BIOS), I have actually hit the desktop at 2.9GHz.
For about 3 seconds.
Then Windows decides to crap it's pants and disappears.
My luck at 2.8GHz hasn't been much better...I can stay in Windows, but any sort of benchmark crashes me out.
2.7GHz seems to be fine, I was Prime stable (for 10 hours) yesterday and temps never broke 30C.
Obviously, this all just for bragging rights really, I'll prolly clock back to 2.4GHz for regular use because I can run that at the tight RAM timings (all of my benches have been run at CAS 2.5, but the RAM is rated for CAS2) AND undervolt the CPU.
I must say that the DFI board is really as good as they say (now I'm one of "they"!) as it has easily surpassed the best settings that the Gigabyte could achieve.
Not to denigrate the Gigabyte...it is a very good board, very fully featured AND it's cheaper, but for sheer performance the DFI is king.
Tomorrow I plan on rerouting the waterloop so the GPU is first in line...hopefully this will drop the vid card temps some.
All in all, both Sprocket and I are quite pleased.
Sooo, how long 'til you buy a dual core then?
Bet it's tempting.
Well yeah, of course it's tempting.
I have been trying to keep up on the reviews, but so far I haven't really seen any convincing proof that the x2's are a "gotta have" item.
It doesn't help that AMD, following their usual assbackwards marketing plan, is releasing the high-end chips first, the more reaonably priced units dribbling behind.
Also keep in mind that sometime after the release of Longhorn (whenever that might be) AMD is planning on jumping platforms...different sockets and probably DDR3 RAM.
Such a move will render my current, (nearly) state-of-the-art setup as technically obsolete as socketA is today and none of my core components will be carried forward.
Besides, I think I finally lucked into a "good'un"...one of those fabled individual chips that surpasses it's brethren.
As far as I can tell, Sprocket's Winchester is one of the best around and easily the equal of many Venice/San Diego's whose results I've seen.
And really, let's be honest...a Duron would handle my needs.
I just do this for fun.
Anyway, my next purchases are going to be the Swiftech Storm G4 waterblock and the Gigabyte RAM drive PCI card with 4GB of RAM.
Then I need the second 6800 GT vid card so I can enable SLI.*
*BTW, I did run Aquamark03 a few times at 2.7GHz.
With the card clocked to 415/1.15 (the numbers that "detect optimal settings" gives me) my best score was 79,488.
I'm hoping to crack 80k if my rerouted waterloop decreases temps enough.
:lol:Quote:
And really, let's be honest...a Duron would handle my needs.
Thanks we enjoy it too.
did you try any performance testing? Aquamark? 3Dmark?
Quote:
Originally Posted by Monkeee
:frusty:Quote:
Originally Posted by clocker
:lol:
(don't mind me, I'm just jinxing a bit.)
@clocker, I wonder if you could get a bit extra if you sold that 3500, what with you having proven it to be a good one :unsure:
How long have you planned on keeping this rig, btw?
Maybe if I sold her now on a forum classified I could get a reasonable price, but as the new chips come out and prices fall, any percieved extra value evaporates.
I never plan anything.
Sprocket has evolved organically, many revisions directly attributable to a stroke of good luck in sales/deals.
It's easier and more fun that way.
Right then class, pencils at the ready.
I've been doing a bit more tinkering with my setup and have made a few improvements along with some surprising steps backwards.
With the Gigabyte board I had struggled mightily to break into the 2500's at PC Pitstop (yeah I know it's crap, but they keep having prizes for high scores and I really wanted another Raptor).
Not a prob anymore...
http://img.photobucket.com/albums/v7...pcpitstop1.jpg
My best Everest memory read test...
http://img.photobucket.com/albums/v7...r/memread5.jpg
Finally, a sooooooo close try at sub-30 seconds at SuperPI...
http://img.photobucket.com/albums/v7...r/superpi5.jpg
WinRAR compression test doesn't seem to respond to anything, I can equal, but not best, my previous score of 38.
Aquamark3 still refuses to yeild a score over 80k.
I'm thinking that my CPU memory controller might now be the bottleneck.
The Patriot RAM has performed fine with the original tight timings ( I only dropped the CAS to 2.5), but relaxing further doesn't help. I can't even POST at cas3.
Maybe there are a whole different set of timings I could adopt to try and break 280 FSB...I'm going to research that.
Right now I've settled back to 2.4GHz and am playing with voltages...specifically- undervolting.
Vcore is down from a high of 1.525 (stock is 1.4) to 1.325v and RAM is at 2.7v.
CPU/GPU temps have budged hardly at all through this trip through GHzland.
I have yet to break 30c on the CPU.
I think that Sprocket has turned out quite healthy for a general-purpose PC.
Compared to hardcore sportclockers she falls short, but she's not as single purpose, stripped down as they, either.
The real players are on machines with SCSI RAID, pared-to-the-bone Win2K and phase-change cooling.
SuperPI scores are nearing 25 seconds at that rarified level.
Of course, after the benchmark they can't browse the forums or burn a DVD like I can, either.
Later today I plan on rerouting the waterloop (which might involve a new reservoir also) and tweaking the cosmetics a bit more...I still haven't decided on what to do with the front bezel.
I can also tell that soon I'll have to deal with the northbridge HSF on this board...it is one whiny little bastard. So far, every socket939 board with PCIe that I've seen has suffered from horrible placement of this chip...effective cooling is an impossibility.
I can imagine some solutions, but none that will allow for the second vidcard should I go SLI. Replacing this unit requires removing the motherboard (which is quite involved for me) so I'd prefer to do it but once.
OK LP, bring it on.....
:blink: I hate you. :dry:Quote:
I think that Sprocket has turned out quite healthy for a general-purpose PC.
Are you getting another raptor then?
http://www.thomasmodels.com/gallery/raptor.jpgQuote:
Originally Posted by SnnY
i dont think it will fit? :wacko: :lol:
Not in a contest, no.Quote:
Originally Posted by SnnY
Hitting 2600 on this test (which two months ago would have struck me as miraculous) doesn't even ante onto the table anymore.
The big dogs have broken 3000...
That shall be my cross to bear.Quote:
Originally Posted by Spicker
In hopes of intensifying Spicker's hatred, here are some more shots.
First, the rerouted loop.
Looks very much like the original, doesn't it?
Yeah, I thought so too.
Trust me though, it's been redone.
http://img.photobucket.com/albums/v7...r/gpufirst.jpg
And a bit of performance data.
http://img.photobucket.com/albums/v7...ker/prime2.jpg
Sprocket had been Priming for maybe 4 hours in this shot.
Oddly, I also had the nvidia temp monitor app open for this shot, but it disappeared during capture.
Dunno why.
It was reading 31C at the time in a room temp of 23-24C.
Soooo.
Rerouting the loop has had zeroapparent effect on the temps.
Neither the GPU nor the CPU went up or down as a result.
Interesting.
As it turns out, yes, I am.Quote:
Originally Posted by SnnY
I didn't think I had snagged it, but it showed up yesterday ($85 off the OCForums classified).
RAID 0, here I come.
After I figure out how to mount more drives in Sprocket.
I have a Seagate 120 SATA that would be nice to install also (for no particular reason other than I have it...) so I need to provide space (and cooling) for two more drives.
Just eyeballing the available space, it looks possible to use the rod/suspension method and squeeze all three drives where previously there was one, but airflow (and thus temps) might be an issue.
Tomorrow is my day off so I'll attack this then.
Ultimately I hope to also incorporate the 4GB RAM drive for the bare OS.
Then the 72GB RAID array would hold the page file and programs and the solo 120 SATA disc would be straight data storage.
This would be about the fastest storage array possible, I think.
So much for the lean, mean, stripped-down Sprocket I originally intended, but as I mentioned previously, serendipity plays a big role in my modding adventures.
BTW, I just heard from FrozenCPU and the Hiper PSU pictured in a previous post should be available in 2-3 weeks.
Right after my next paycheck.
Oh oh.
Sorry to interrupt but there's a hawt topic on /. and I thought maybe you guys like to take a peek also.
(I know you peps like this stuff)
For hardcore serverpr0n, check the:
Homebuilt 19" Mini-ITX Server Rack.
:D
Oh, that's cruel...link is down.
Was it from here: http://www.mini-itx.com/ ?Quote:
Originally Posted by Afronaut
There's one among the projects, anyway.
Somehow I did manage to feck teh link.
here it is:
http://home19.inet.tele.dk/jys0500/rack/
the slashdot link is in my earlier post as ---> /. but here is that too:
http://hardware.slashdot.org/article...id=222&tid=137
Come to think of it, I dont think it has the latest hardware but me been a
some sorta musician I get woody just seeing the rack, hehe...
:D
Sprocket is now RAIDed.
It was a stone bitch to achieve...DFI provides zero documentation and the required settings are spread out all over the BIOS.
Literally just finished the install as I flew out the door for work this AM so there is no performance data as yet.
Shall spend some time doing the normal XP tweaking and loading and see whether this nightmare was worth it.
Hmmm.
Well, is RAID worth the hassle to set up?
You decide.
Here is a shot from a few months ago.
This is a single Raptor...
http://img.photobucket.com/albums/v7...er/raptor2.jpg
Now the Raptor is joined by it's twin in RAID0...
http://img.photobucket.com/albums/v7...r/raidtach.jpg
Rather dramatic difference I would say.
Is the drive access speed noticeable when using the PC?
Can't really say yet, VB.
Haven't really had a chance to do anything, still installing and tweaking XP so regular use hasn't happened.
It seems likely that there should be a noticable difference, but I'll let you know.
They don't support NCQ too, do they? :unsure:
No.
Raptor's make up with sheer speed their lack of new tech.
New record for SuperPi...
http://img.photobucket.com/albums/v7...r/superpi4.jpg
I shall break into the 29's with this rig.
FYI, the lowest score I've ever seen is 26.2 seconds.
Don't think I can get there, but it's fun to try.
I thought as much, I wonder how fast a raptor with ncq would go.
(congratz on the speeds, btw)
You've certainly achieved some impressive gains in both sequential read speed and burst transfer speed. Unfortunately, since we rarely have control of how our data is written to the disk large scale sequential reading is so rare that it is unlikely to be a factor in day to day performance. Sadly most tests seem to want to show raw power rather than usability.
A better test would have been a few minutes of totally random access, but unless you are going to de-raid for comparison it is rather meaningless. And of course since you have 2 disks it would be meaningful to do a similar test with both the drives being accessed randomly in non-raid mode too.
The latest 74GB Raptors certainly support NCQ, but I don't know about the ones you've got (though documantation would suggest not). I wonder how much difference that would make, especially in non-raid mode.
You certainly make valid points...clearly, all this benchmarking is just for bragging rights, not for day-to-day usability.
And speaking of bragging....
http://img.photobucket.com/albums/v7...r/winRAR40.jpg
If I can break past 41 I'll hit the top three in the competition.
Anyone know what winRAR is most dependent on?
Have you read the help file?
Apparently it doesn't start showing meaningful data until at least 10MB of data has been processed. In theory that means that the amount of data processed in 60 seconds is not actually significant, it is the resulting speed which is important. Having said that, multiplying your resulting speed by 60 does actually give 40MB (allowing for rounding).
Apparently the program generates random data, then compresses it and decompresses it simultaneously. As far as I can tell there is no actual disk access involved. By way of confirmation of that, a virus scan running at the same time hardly seemed to affect the performance.
There seem to be three processes involved:
Generation of random data.
Compression of that data.
Decompression of the compressed data.
There is also comparison of the original and resulting data streams, but I don't think that would have much bearing on the result.
Obviously, since data is constantly passing from the processor to memory, the memory management system must be highly involved.
I can see little advantage (and many disadvantages) in the software using floating point arithmetic for the compression and decompression but I suppose that depends on whether the software designers were smart enough to realise that. I would guess they were.
All other filters and algorithms are disabled, so the only other factor is pure processor power.
All in all I would say that the benchmark tests a mixture of memory access speed and processor performance. The one thing I can't tell from your images is the speed the memory is running at, so I suppose it is 260MHz.
Just found out that revision E chips (Venice, San Diego and x2) have hidden memory speed settings as long as the bios supports it, namely 433, 466 and 500. That'll upset all those people who think AMD are stuck at DDR400. Of course if, as is likely, those speeds are achieved by a multiplier of the base frequency then 500 would become 650 with your 260MHz base. Bet you wish you'd held out for the Venice core now. :P
The last couple of Oskar Wu beta BIOS releases have had the memory dividers in place as you describe.
I had never seen a "plus side" divider before (i.e., over 1:1 favoring the RAM) and naturally, HAD to see what happened.
Well, long story short, you gotta have an E-revision chip to make that work.
It wasn't pretty.
Women ran screaming and strong men lay twitching on the floor.
Sprocket was bleeding from her IDE ports.
I've reflashed to the June official BIOS release and may even go further back.
The last several betas have all addressed issues involving the x2s and the e's in general...much of which might actually be detrimental to Winnie performance.
http://img.photobucket.com/albums/v7...r/winRAR41.jpg
http://img.photobucket.com/albums/v78/clocker/281.jpg
Two firsts in a row.
Although I've booted at 2.8GHz before it was never stable enough to even get a screenie.
I could do some stuff this time although SuperPi crapped out before completing one iteration.
Back to work.
My, but Sprocket has been feeling frisky this morning.
http://img.photobucket.com/albums/v7...r/winrar44.jpg
And...
http://img.photobucket.com/albums/v7...er/sub30pi.jpg
I think a polite "w00t!" is in order.
Polite w00t!
:D
I think you mean w00t!Quote:
Originally Posted by Afronaut
:D Another W00t! from me.
Aw shucks.
Sprocket is blushing...
http://www.movie-winners.com/posters...newithwind.jpg
:dry: I hate u :dry:Quote:
Originally Posted by clocker