This is excellent work nik as usual. It had to be very tedious with that many blocks x mounts and test data. :clap: :clap:
The test done with a DDC and a D5 would be very interesting to compare blocks performance.
Printable View
This is excellent work nik as usual. It had to be very tedious with that many blocks x mounts and test data. :clap: :clap:
The test done with a DDC and a D5 would be very interesting to compare blocks performance.
I wonder if these nozzles would give the same performance gain with a Quad Core?
Wow, that's how testing should be done. Kudos to all your hard work, it is very much appreciated!:up: :up:
crap:bsod: ..figures..I go and buy a GTX and low and behold a Fuzion performs better at least on dual core...quad core will be interesting..wonder if the larger die will allow a bowed GTX to equal the Fuzion(bowed with 4.5 nozzle)
crap:bsod: ..figures..I go and buy a GTX and low and behold a Fuzion performs better at least on dual core...quad core will be interesting..wonder if the larger die will allow a bowed GTX to equal the Fuzion(bowed with 4.5 nozzle)
Great work and thanks for your efforts
BTW...lucky dog getting at least a G7 prototype to try:slobber:
Nice testing, keep up the good work :)
Thank you for this work. It's greatly appreciated.
great worklog...
+1 :up:Quote:
Originally Posted by nealh
Bump for the new graph in the first post.
Good to see affect of mount exemplified - good / bad mount can account for upto 2.5 degree swing in temps in either direction... effectively negating benefits of one block over another... ie: a good mount on a bowed GTX would beat a bad mount on a nozzle-bowed FuZion and viceversa... average difference between the two assuming average mounts allround is approx 1 degree in the FuZion's favour...
Actually no, even the worst mount of the bowed/nozzle fuzion recorded better than the best mount of the bowed GTX, of course not by much at all.Quote:
Originally Posted by Marci
One thing to note, bowed bases tend to be more inconsistent in regards to mount variation - it should be obvious as to why.
Depends whether you consider 5 mounts to be "enough". Could be you had one REALLY bad GTX mounts, and 4 reasonable mounts, but no REALLY good mounts, whereas with the fuzion bowed with nozzle you got one REALLY good mount, and 4 reasonable mounts, and no really bad mounts... if u get my drift? Is the sample size enough to cover the potential mounting variances...
I'd like to see (entirely up to you of course) a lengthier set of testing concentrating on these two blocks (and poss G7 if it falls near / within these two in the overall mix), consisting of 10 to 20 mounts possibly... just to show max possible variance between mounts (ignore flat bases as it's much easier to get good mount with them)... use torque wrench with GatorGrip and replace thumbscrew with a suitable nut to get a representation of mounting torque? Or are we heading too far down ProC roads...?
Marci what you propose would not be fun :D I believe that with the 5 mounts a trend has been established... I also think that if we performed 100 mounts of each the current trend with the 5 mounts would continue. Maybe if I get bored...
Heh... definitely not fun... no
Bring on a bowed Fuzion with nozzle :)
Went to machinists yesterday, and asked to get a new top made up for the incomplete G7. They had apparently backed up all my block data onto a CD, but something had gone wrong, and the CD didn't have my data on it, and all the original mastercam plans were gone.
i.e. the machinists have lost all the Mastercam work for all of my blocks and we're back to the paper plans again. :rolleyes:
Oh well, what they lost wasn't like to get made again anyway, but it'll take a couple of extra weeks to get the G7 top made up.
poop... a lot of work that mastercam is... well... coudl be a LOT worse... lolQuote:
Originally Posted by Cathar
is this the o-ring used to bow the fuzion?
http://www.petrastechshop.com/ekthspforg1f.html
Very methodical testing nik...obviously a great expenditure of your time and effort. Test bench looks very slick too. :up:
Is the D-Tek nozzle for sale ?
Not in shops yet, as they say.
Marci
Racing sidecars is fun.
Slaving over multiple block mounts is purgatory.
Don't you like Nik'?
:rofl:
im willing to bet the data to be different on a quadcore.
i wish you had one to test on. :\
Reason why im saying this is becuase of core locaiton and the fact a quadcore is dual die stacked next to each other. There is a much larger cooling area required on the Quads vs a Dual Core.
This is also the reason why i think a block thats ment to cool a 52mm TEC would perform better. The GTX can support a 52mm TEC, while the Fusion cant.
So if I knock 5º off these numbers I'd have a ball park predictor of these blocks on a E6600 with a PA120.3 loop?
The only info I've for comparing HE and PA120.3 seems to point to about five degrees performance difference.
You will have an answer for that soon. Fuzion will support quad cores with no problems (and I am no fanboy of D-Tek or Swiftech).Quote:
Originally Posted by NaeKuh
Jay