Forum Home
Press F1
Thread ID: 84104	2007-10-24 10:15:00	Thermal Paste?	bigmit37 (7537)	Press F1

Post ID	Timestamp	Content	User
604904	2007-10-26 19:43:00	The author of the article makes it very clear at the beginning that performance over time should be the focus of any assessment of thermal pastes - and then goes on to test everything for, at most, a day. That's why tap water performed well - until it evaporated. After applying Arctic Silver 5 to my E6750 I watched my temps. True to the manufacturers claims, after about 200 hours (probably less) the temps dropped by around 5 degrees C. This is temperatures reached during long runs of Orthos stress test. His claims with regards to the gains of a small tweak are a little narrow in their focus too, because a number of small tweaks can add up to a significant gain. Also, "real-world gains" are not necessarily the focus of the enthusiast overclocker. They're usually interested in benchmarking for the sheer sake of benchmarking. It may seem pointless to some - but then golf seems pretty pointless to some people too.	Deane F (8204)
604905	2007-10-26 20:01:00	After applying Arctic Silver 5 to my E6750 I watched my temps. True to the manufacturers claims, after about 200 hours (probably less) the temps dropped by around 5 degrees C. This is temperatures reached during long runs of Orthos stress test. 5 degrees C is a lot of "improvement", was it achieved because the initial results weren't as good as expected?	PaulD (232)
604906	2007-10-26 20:27:00	5 degrees C is a lot of "improvement", was it achieved because the initial results weren't as good as expected? I started out with the interface material that was on the stock Intel boxed HSF - but I had motherboard problems so when I got a replacement motherboard I had to remove the Intel goop and reapply with aftermarket paste. Initially, the Arctic Silver 5 showed exactly the same temeratures as teh stock Intel stuff, under full load during Orthos stress test (in which the maximum temperature plateaus after an hour or so) Max temperatures of 48 - 49 degrees C. After about a week and a half - having run Orthos a few times - I noticed a drop of the max temp down to 43 degrees during the stress test. So, it's only 5 degrees downward shift - but I have been able now to use a fan controller to turn down my case fans (particularly the intake at the front) so my system is now more quiet for the same CPU core temperature as before. Another thing the author doesn't mention is the effect of temperature on CPU life - the rule of thumb being that every raise of 10 degrees halves the life of the CPU.	Deane F (8204)
604907	2007-10-26 21:22:00	Another thing the author doesn't mention...Keeping in mind that Dan doesn't do thermal paste testing day in, day out as a job. I'm sure that dedicated overclockers could go to quite alarming degrees of fussiness. And remember he does take the piss a bit - note some of his terminology. I think his comparisons were excellent for the masses. the rule of thumb being that every raise of 10 degrees halves the life of the CPU.To be honest I think that's a fallacy.	Greg (193)
604908	2007-10-27 05:44:00	i heard 10 degrees increase/decrease was 10% performance decrease/increase	motorbyclist (188)
604909	2007-10-27 06:18:00	Any reduction in cpu temperature with time using Arctic Silver compound can only arise from thinning of the grease film due to the constant pressure applied, what in lubrication circles would be called "squeeze film" effects. It is amenable to calculation. It implies the grease film was too thickly applied to start with. There was a long posting on these topics 3 years ago: pressf1.pcworld.co.nz The image files in that thread are no longer are dead links, but I can probably find the pics and repost on ImageF1 if anyone is interested.	Terry Porritt (14)
604910	2007-10-29 00:44:00	"It's got the same advanced polysynthetic nuclear fish discotheque technology" Great quote, could also apply to many other products (eg audio cables) "Is fancy thermal goop as good as the manufacturers would like you to believe, and as various Junior Overclockers do believe? Nope. Is it, therefore, a complete waste of money? Absolutely not." "If your PC's so marginal that a CPU running three or four degrees Celsius warmer will crash it, the solution is not to try to edge away from the precipice with better thermal compound - it's to make a big change to the cooling system, or just slow the darn thing down." Ive been using generic silicon paste for years, works fine. If you want to really lower CPU temps, get a temp controlled room(or put it in a fridge:2cents: )	steveroby (9470)
604911	2007-10-29 01:15:00	I have always used the ceramic grease from Artic, purely because its so much nicer to work with, and cleaner. I have 2 tubes of it, They have lasted me for nearly 5 years and more cpu's then you could chuck a muddy stick at.	Metla (12)
1 2