I rent.

Update 9-23-03

Here's my latest iteration of a quiet PC as of 9-23-03.

I was getting bored with my P4 2.4C rig so I decided to try a 3.0C CPU and a new Asus P4P800-Deluxe MoBo. My prior Intel 875PBZ board was stable but Intel's BIOS releases for that board have been disastrous. The last 7 releases have (among other things) disabled the PAT and caused cold boot/reboot and BSOD issues.

Latest Hardware is:

• Asus P4P800-Deluxe with "PAT" enabled in BIOS.
  
• P4 3.0C CPU
  
• Thermalright SLK900U heatsink + 92mm Panaflo L1A "BX" rpm monitoring fan @ 9 volts via Zalman Fanmate **Had to speed up fan from 1500rpm to 1650rpm with faster, hotter CPU**
  
• 2 x 512MB of Mushkin Level II Dual Pack PC3200 RAM @ 2-3-3-6, 2.65V **Will run stable at 2-2-2-6 with 2.75V**
  
• ATI Radeon 8500, fanless
  
• Seagate Barracuda IV, 80GB
  
• Plextor Premium 52x CDrw **possibly the quietest CDrom/CDrw drive I've ever used**
  
• Seasonic Super Silencer 400w PSU **Noticeably quieter than the Nexus3000 that it replaced, PSU is no longer the loudest component in my system**
  
• Antec SLK3700 case with 120mm "OEM" Panaflo @ 5 volts as case exhaust fan, 80mm L1A @ 6-ish volts as HDD cooling fan

Pics will follow later.


Here's my temps with the 3.0C replacing the 2.4C:

Ambient temps in my computer room range from the mid 70°F to mid 90°F range during this time of the year. All temps will reflect this spread.

• Mobo temp ranges from 30°C to 41°C at idle, 31°C to 45°C at load.
  
• CPU temp ranges from 33°C to 39°C at idle, 47°C to 59°C at Prime95 load.

These are perfectly safe temps and this computer runs super quiet at these settings. Even when the temps are at their hottest, the PSU fan does not increase it's speed.

The swap from the 2.4C to the 3.0C required that I increase the speed of my 92mm L1A(BX) CPU fan from 1500rpm to about 1650rpm and temps are still around 3°C hotter at idle and 6-8°C hotter at load. You can now hear the CPU fan over anything else in the case but just barely and only when the ambient noise level is very quiet. I suspect when it cools down after summer I'll be able to turn the speed of the CPU fan down a few hundred rpm and be back at my prior noise level.

Sounds good Ralf. Glad you mentioned the BIOS issues with the Intel board. I'm not sure if my next board will be Intel or Asus but I'll keep this info in mind.

Do you know much about the 865PERL in terms of BIOS issues?

I think the PERL may be OK. I've only heard of one guy having a problem with it but what worries me is that he was having one of the same exact BIOS-related BSODs as the 875PBZ guys.

It may be that there's less complaints about the 865 board because fewer people are using them, or it may actually be a typical Intel board, i.e.: PnP & 100% stable so you never hear anything about it.

Well thanks Ralf for the info. I'm a little disapointed to hear about the Intel board. It seems like there is no 100% trust worthy company these days. They all seem to have their shit products, some just more than others.

THANKS RALF! You are a pro!

I had never seen these sites,
Best,
tk

Impressive system. I have a similar one, but I am still on the 2.4 Ghz. The biggest change for me was switching to the Seasonic PS, like you suggested. That thing is amazing.

I am just about to replace the Cuda with a Samsumg 160 GByte so we will see what that does for the noise, and the performance.

jerry

Lately I've taken another fork in the road. I built another system around my leftover 2.4C processor using an Albatron 845PEV Pro-800 mobo that is undervoltable. It's an 845PE chipset so it supports hyperthreading but no dual-channel RAM but it also has an 800MHz FSB and supports DDR400 RAM.

What I've done so far is put the system together with the 2.4C running at 800FSB and the (2 x 256MB) RAM running at DDR400. The default Vcore for this CPU is 1.525V and when I tested the temps at default Vcore I got 35°C at idle and 53°C at 2xPrime95 load. That's at 76°F ambient with a SLK900U + 92mm L1A running at 1400rpm.

I then undervolted the CPU to 1.25V and my idle temp went down maybe 1°C but my full load temp dropped to 40°C!! That's a 13°C drop!!

I've been running Prime95 with this configuration for about 12hours so far to make sure it's stable. If it makes it to 24hrs without crapping out I'll be a happy camper and I'll run Memtest86 for another 24hrs to make sure everything's OK with the memory subsystem and that'll be that.

What am I going to do with it when I'm finished? Dunno. HTPC maybe?

Dangit Ralf... I know you've mentioned before WHERE you are in SoCal, but it can't be too far from Ventura. I'm gonna have to see (and try to hear) these systems some weekend.

Yep, like I've been saying: undervolting is hugely important, if you're serious about a quiet (performance) PC.

I had the same (limited experience) impression of Intel boards that Ralf had. I'm interested to hear what people say about Tyan since my friend and I swear by them. They have been rock solid for us. My spankin' new Tyan Trinity i875P (S5101) ANNRF is DA BOMB! I still have to do some tweaking around with it to test it out but I'll let you know how it goes on my thread in the newcomers section.

By the way my new system is quite similar to Ralf's. I don't think I'm the only one that copied him though.

I've been waiting to hear an update about your experiences with that Tyan board. Please post something when you get it shook out. One specific question I have is: does it undervolt?

Ralf,

Did you cut a 92mm hole in the front stamped grill or did you leave it alone?

I left it alone.

I meant to ask you why you did that, but then I got my AMB from the store and I can see now that the front grill is less restrictive than the back one.

Is it the reason?

Is it the only reason?

No, the front grill is not less restrictive than the back, the hole size and spacing is exactly the same for both grills.

I left it alone for 2 reasons:

1) I had hoped I would need no front fan at all because it looked like the case had very good ventilation. So I was hoping to save my hands from having to cut out 2 sets of fan grills with the hand operated nibbler.

2) If I did need a fan, I figured it would only be needed for spot-cooling of the HDD and I've found that a 92mm L1A running at 5 volts is the best compromise of low noise vs. airflow. Since I'd need a place to mount it and that wouldn't exist if I cut a 120mm circle out of the case I left the grill in place and drilled a set of 92mm mounting holes and a set of 80mm mounting holes. That way I could mount any fan I wanted.

In fact I did not need any more case ventilation but I did need spot HDD cooling so I did mount the 92mm L1A on the front of the case.

If I was starting out with a brand new SLK3700AMB I would still do it the same way.

very nice job!

Good job, Ralf. How much did your new system cost?

And say, you placed your heatsink fan backwards so that it sucks air instead of blow air. Do you find that it cools more efficiently this way as well as reduce the noise?

Beats me. Add up the prices fron Newegg, that'd be in the ballpark.



I didn't put my fan on backwards. It's in the conventional "blowing down on the CPU" orientation.

Just FYI, heres another batch of 40...

http://cgi.ebay.com/ws/eBayISAPI.dll?Vi ... egory=3673

Oh how I hate this sentence: "Shipping is limited to addresses within the United States (U .S.) ONLY."

In Denmark we have absolutely no wholesale options for these things, and a similar quality 120mm fan would run you upwards of $20 a pop.

GRRRRRRRR

Well, I'm thinking about coping Ralf's design there and I need one of those fans. If people want, I'll pick up the lot and I'll sell them for cost plus shipping. Let me know if anyone would be interested.

Schloktins

I think there's someone here who bought 40 of them and still has a bunch left. Search for "oem AND panaflo" (or some derivation of that) and see what you get. You might look through the threads in the Classfieds forums too.

It's Tom Brown in this thread.

The OEM panaflo is rated to 0.2A; the retail one at 0.24A - this suggests that the OEM is a less powerful version. Maybe the blade design makes up for it but I doubt it. The OEM version probably has an inferior airflow.

That's kind of a rash statement based simply on the Amperage and no practical knowledge. I use the OEM Panaflo and I have found it to be superior to the L1A, in noise and in effectiveness. The design of the blades is intended to create a "tornado" effect, so that the fans works better in contained environs, and can concentrate the airflow to help push through obstructions (grills, heatsinks, radiators, etc.)

Based on considerable firsthand experience with multiple samples of ech type of Panaflo I totally agree.

That is why I used the word "maybe" in my sentence

It was simply a suggestion.

Does the guy still have OEM Panaflo's left?

Contact Tom Brown (see this post) Sounds like he's not too thrilled with these OEM Panaflos and has a bunch left over. Maybe he'll sell some.

Ralf, about the way you mount 92mm Panaflo fan, from the picture it certainly looks like its sucking rather than blowing.