Thermaltake Fanless103 for AMD64?

gotensan01 · Post by **gotensan01** » Fri Jan 21, 2005 11:06 pm

Sorry if this is a repost.

http://www.newegg.com/app/ViewProductDe ... 047&depa=1

Above is a link to a Thermaltake Fanless103 heatsink that I am heavily considering for my 3200 AMD64. I just wanted to know if anyone has used this before or has heard anything about it. I'm not looking to overclock, and temps can be a little high for the sound trade off. I just think it would be nice to not only have a completely silent CPU cooler, but also it would be nice not to have to worry about a fan malfunctioning.

Tibors · Post by **Tibors** » Sat Jan 22, 2005 4:26 am

If I plug "Thermaltake 103" into the search page (don't forget to click on the "Search for all terms" radio button), then I currently get 10 matches.

The thread called anybody seen this? maybe not all thermaltake stuff is crap contains the most usefull information. The bottom line of that discussion: With the right setup, this can be an adequate cooler.

GrahamGarside · Post by **GrahamGarside** » Sun Jan 23, 2005 12:49 pm

if you don't want to worry about the fan malfunctioning go intel, you can remove the entire heatsink from their chips and no burn them
thermaltake make reasonable budget products, in fact their copper heatsinks are quiet good if you were to lap the base
but their concept of quiet is baffling, and to be frank quite often just a lie
they make claims like 0db for their passive cooled psu's, and most of their claims for fan noise are just not true
I personaly wouldn't buy thermaltake products for the purposes of building a quiet system. Now if I were on a budget and wanted reasonable cooling, where the noise level wan't a huge issue then I would consider them

meglamaniac · Post by **meglamaniac** » Sun Jan 23, 2005 3:02 pm

GrahamGarside wrote:if you don't want to worry about the fan malfunctioning go intel, you can remove the entire heatsink from their chips and no burn them

That's because Intel thermally throttle their CPUs. So yes you can run it without a heat, but nowhere near the speed it would run at with one.

GrahamGarside wrote:thermaltake make reasonable budget products, in fact their copper heatsinks are quiet good if you were to lap the base
but their concept of quiet is baffling, and to be frank quite often just a lie
they make claims like 0db for their passive cooled psu's, and most of their claims for fan noise are just not true

Ehh.. passive PSU's don't make any noise other than maybe coil whine or buzz, which you'd be hard put to get to register on most sound meters, so actually 0db is correct (as it is for all passively cooled products).
I agree with the fan part though.

Overall I'd still go with the thermaltake XP120 - you can get excellent results with the fan at 5v (basicly inaudible) and the option is there to ramp it up if required. I wouldn't trust a totally passive cooler unless I had a lot of airflow through the case, in which case that'd defeat the point of the silent cooler...

ronrem · Post by **ronrem** » Tue Feb 01, 2005 8:04 pm

Coolmax has a PSU with a giant 140mm fan that is a 3 speed with the slow at,as I recall,less than 1000 rpm,with minimal noise. The Thermaltake passive i saw is a heatpipe type and assumes some case fan provides adequate airflow. Even spinning slow the 140 may move air enough to cool the PSU and pull air through the heatpipe/heatsink. To be prudent I likely would have a second fan,maybe a quiet 80,maybe a variable that could be a safety margin if weather is warm-but could be shut down. Assuming quieted Spinrights and a single ,big,Fan at 16-18db or less the internal noise could be quite low and so the case could be set up more for good airflow than insulation

GrahamGarside · Post by **GrahamGarside** » Wed Feb 02, 2005 10:58 am

yes but a cpu throttling is preferable to burning up, if you really have concerns about stability and fan failure then intel is the best choice. I'm no intel fan I use an athlon64 but in crucial situations I'd count on an intel.

and passive psu's do make noise, you said yourself coil buzz, it may not register on a cheap sound meter but it still making noise. And the actual convection of heat it's self away from the psu is technically making noise. Our hearing is finely tuned to the point that our sensativity is just above the point of picking up convection in the air so we wouldn't be able to hear it, but my point was only that you can tell how reliable thermaltake are from this claim of no noise

meglamaniac · Post by **meglamaniac** » Wed Feb 02, 2005 11:18 am

Well, I think counting air convection is taking it a bit far, even for SPCR!
I'm fine with the definition of a passive PSU as noiseless. It's not just TT who use that definition.

As I said though, I agree that their other products are far from accurate in terms of noise ratings.

GrahamGarside · Post by **GrahamGarside** » Wed Feb 02, 2005 2:21 pm

I'm not even getting at the fact that they call it noiseless, it's just the unscientific way they make the claim, 0db is ridiculous, you can't make that claim

JanW · Post by **JanW** » Wed Feb 02, 2005 4:27 pm

Someone correct me if I'm wrong, but the db scale being logarithmic, there is no (finite) absolute zero on that scale. IIRC 0db is defined as inaudible to the human ear, but technically speaking you could talk about something making -3db noise (less than the threshold of audibility, convection for example).

meglamaniac · Post by **meglamaniac** » Wed Feb 02, 2005 11:01 pm

Yeah this is something I've never quite got, I don't think I understand dB ratings all that well.
For example, in Adobe Audition (audio editor) you can set the vertical scale (which measures amplitude) to dB - and it runs from negative infinity at the centre where it's quietest to 0dB at the outside where it's loudest.
I usually have it set to Normalized Values instead which make more sense to me (0.0 to 1.0).

Hmmm.

JanW · Post by **JanW** » Thu Feb 03, 2005 1:17 am

Ok, I did a little reading, and here it is:

Decibel by itself is actually quite generic. Strictly speaking, it's not a unit of measure, it's just an indication that a quantity is on a logarithmic scale (based on the decadic logarithm). The usual use is to compare two quantities of the same units (preferentially power levels) on a logarithmic scale. The power level P1 relative to P0 expressed in dB is:
M = log10(P1/P0) Bel = 10*log10(P1/P0) decibel.
0 dB means, both quantities are equal:
log10(P1/P0)=0 => P1=P0.
If P1<P0, then the ratio is smaller than one and the logarithm is negative.

In Adobe Audition, the reference level P0 is the maximum that is achievable (there is no other reference that would be reasonable), so that attenuation is shown (by definition always smaller than 1). You'll also find this use on the volume dial of some HiFi amplifiers, where the maximum volume is marked 0dB and the lowest might be -infinity (complete attenuation).

When sound pressure (I use lowercase p for pressure) is measured (or voltage, or current) there is a little twist here, since power (uppercase P) depends on the square of these quantities. So
M = 10*log10(P1/P0) dB = 10*log10(p1^2/p0^2) dB = 20*log10(p1/p0) dB.
To express an absolute sound pressure level, a reference of
p0= 0.0002µbar = 0.00002Pa
is used. So 0dB of sound pressure is this reference level. Any sound pressure level lower than this would mean negative dB.

Plugging all of this back in, if you have a sound pressure level (SPL) expressed in dB, the absolute sound pressure in Pascal is:
p1 = 0.00002Pa * 10^(SPL/20)

Hope any of this makes sense to someone....

EDIT: 'ed serveal times to get units straight... Someone please correct me if I got it still wrong