Printable View
I just know someone is going to mention how maybe the block wasn't mounted well, in either gabes case with the storm, or orkans case with the apogee, this debate will never end until a few independant testers can get thier hands on each block and do the tests. Its been quite entertaining i gotta say!! :rofl:
nice... I hope my Apogee does as well tomorrow...
Silliness?Quote:
Originally Posted by MaxxxRacer
We are not arguing that different CPU's of the same make and model will overclock at completely different levels.
We are arguing about the scale of the report. 100Mhz (3% OC in this case) is a huge difference, you know that as well as I do. We are not trying to beat Orkan's numbers, we are trying to see if there is a difference in overclockability between these two blocks. Read my latest post to see our results. So far, we see no difference whatsoever.
Another note, to ALL:
I run a business and work 6 x 12. Unfortunately I cannot spend much time on these forums. I spent part of last sunday answering some posts, but please forgive me if I simply cannot address all your questions.
Gabe: I'll pass your results around a few forums as alot of people would like to see your findings im sure...
I, for one, believe everyone's results when it comes to overclocking tests.Quote:
Originally Posted by Hu1kamania
Has been seen many times before, but eva2000 researched it quite a bit and took a lot of time. Across a range of IHS capped CPU's, put block A on, and measure the overclocks across a range of CPU's. Put block B on, and the overclocks are different, sometimes ranging up to 100MHz difference, but not in all case is one block necessarily better than the other. CPU 1 may very well work well with Block B, and CPU 2 may very well work better with Block A. This has been observed by more than just eva2000 for those who have taken the time to do so.
The pattern is quite clear. die->IHS->wb interaction is unique and specific for each individual processor. It is not something that can be generalised on with a sample size of one. Would probably need a sample size of maybe 20 to 100 CPU's to form any "trend" with regards to differences between two blocks. This is the specific problem that I have always had with IHS testing, and I noticed it 3 years ago when I started testing on the IHS capped P4's. Since that time, there has been a gradually increasing body of evidence to support the statement that when it comes to an IHS bound CPU, what block works best with that CPU is near totally random.
Pull off the IHS, and all the inconsistencies go away (after multiple remounts to qualify the die->tim variations).
Additionally, there's always the issue of whether or not the motherboard, power supply, memory, chipset, mobo power regulation, or whatever, is merely flat-lining the overclock. Overclock flat-lining is quite a common problem, and it takes a lot of effort to find and build a system where particular sub-components are not limiting a CPU's overclock independently of whatever the cooling is doing. This is an issue that applies to all CPU's though, whether or not they are bare-die, or IHS capped.
There's no right or wrong here, nor is there even a valid disagreement, not when the body of evidence for the inconsistencies for IHS use is ever growing.
I trust his results. I was just pointing out how this debate seems endless as someone is going to find fault in someone's testing methods and invalidate them. Invalidate to them self, not necessarily to everyone else. Its a constant game of creating doubt, and it makes me smirk.
Not enough time, I am like the rabbit in Alice.. I'm late, I'm late.. Seriously, I am fully aware that these are "quick n dirty" tests. And they were done partly to inform users in this forum, but also for my own personal peace of mind. We had not done any OC in our battery of tests, and that was a mistake. It is now corrected. There are more independant articles to be published pitting the Apogee vs the Storm, they will present more data hopefully validating our own tests.Quote:
Originally Posted by dinos22
Some people are interpreting or trying to build all kind of conspiracy theories in this forum and others, and in absolute honesty there are none. We are indeed using the Storm as a benchmark to validate the Apogee. The fact that this is rubbing some feathers the wrong way is irrelevant to us. Some wonder why not let the chips fall and let the market decide? We are. The chips will fall all by themselves. Truth ALWAYS comes out. Meanwhile, we simply address the issues that we perceive wherever and whenever we see them. All we are doing is to present a complete picture, and let the people decide. I will add one last thing: we do believe in our IP :-) But is the present embodiment of such IP the apogee of water-cooling technology (pun intended) I certainly hope not.
Sounds like you need a vacation gabe. I hope you get some much needed rest during the holidays! :toast:Quote:
Originally Posted by gabe
the 100Mhz number is not relevant in itself. The frequency scale to which it applies is.Quote:
Originally Posted by Cathar
gabe let us know how your internal testing is going for the Apogee suspect crack area on orkan's block and your own "off the shelf" blocks..........I would still be reserved in buying one of these blocks if it would crack under pressure.....wouldn't want to shell out a few more $1000 for my PC gear.....i understand that it is one example (i.e. orkan's block) and that robotech mentioned that as an issue in his review......but still it makes me uneasy
The block top should not be an issue...
Gabe, have you ever tried that CPU under anything really cold? it could very well wall at that frequency; so could the motherboard. just curious. Wanna give it a crack with an old school block (ie maze3 or similar) for an interesting comparison?
All CPUs are different... i believe every OCing result comparison posted in this thread.
I have already, in previous posts. Stress tests are negative. Specs wise, it's a non issue. QC has been briefed and warned: zero tolerance.Quote:
Originally Posted by dinos22
i thought your previous post was saying that you were in the middle of testingQuote:
Originally Posted by gabe
that's cool.....i don't know what all the fuss is about.....the results certainly don't show any dominance by Storm as of yet.......let's see what happens with other users and yeah by the time everyone is done we'll have more than enough examples to conclude how apogee vs storm will end up so i wouldn't get to caught up in testing validity argument....numbers will prove it
it would be interesting to see whether MCW6002 block fairs well against these....wonder if eva2000 will jump onboard here as he's already an owner of a few G4s and 6002 as well as other hardware....not to mentioned all the numerous CPUs he can test on
Gabe, you return and post several times... yet you still refuse to answer my question:
Why are so quick to badmouth the storm, in apogee's stead, when it is also one of your products?
In my entire professional career, I have never witnessed a company attempt to "slay" one of their products like you are trying to do with the storm. It is quite apparent that no one is going to be able to figure anything out on this for quite some time.
I had some time today, so I swapped the apogee in my loop in place of the WhiteWater on my newly stripped X2. Boy oh BOY does it tell an interesting story!
The people that aren't idiots... already know the results of that test.
Waterblock performance? This is not the issue at hand. The issues at hand are simply this:
1) marketing
2) money
3) IHS
A) Do I believe the average user with an IHS will see a difference between the storm and apogee? ... or the Apogee and a Spiral for that matter? No, probably not. The storm is the only block I've been able to hit 2.8ghz with... but hey, maybe it is just my cpu... or a freak of nature.
B) Will the enthusiasts that choose to remove their IHS see a difference? ...lol I thank those of you that tried to point this out to me before I stripped mine.
C) I am exiting this "argument" as I feel no ground will be gained on either side for quite some time. My own personal results are satisfaction enough.
---------------Bottom Line---------------
If you have an IHS on your processor - Buy an apogee, or a maze3, or a spiral, or whatever block you want, as it probably won't matter at all. As many tests and results show, they will be within 1 or 2c of eachother.
If you DONT have an IHS - Buy a storm... or whatever Cathar's next block is. ;)
alright orkan why don't you post your actual results with IHS-less CPU.......somehow i doubt there will be much of a difference without IHS but hey I could be wrong=speculation(bottom line)..........show us the results pls
how is gabe slaying Storm again? I must have missed that bit somehow:confused:
One can safely assume that when talking about A64's, we're generally talking about a 100MHz difference in the 2.5->3.0GHz frequency range.Quote:
Originally Posted by gabe
Interesting. I guess I always assumed that with a name of George, that eva2000 was male.Quote:
Originally Posted by KaptCrunch
i'm waiting for Catha's next block :)Quote:
Originally Posted by orkan
me too heheheh.....maybe George had a sex change heheh :cool:Quote:
Originally Posted by Cathar
:rolleyes: :stick: :D :DQuote:
Originally Posted by dinos22
All I know is that my next proc (hopefully a naked 170, but in the meantime probably a naked winch) is gonna have a storm shooting little jets of water right on top of it.
You haven't even tried the Storm yet without the IHS have you? Dude, I even agree with you for the most part but you sure make it hard to do with all the added BS.Quote:
Originally Posted by orkan
-pickles
Orkan, I've been as skeptical as you about this whole deal, and curious about the reasoning behind it, but I don't think it's reasonable to expect any company to dish on its internal business strategies.
Disclosing performance data and testing methodology is one thing -- not to mention necessary in a market dominated by hardcore enthusiasts -- but revealing purely business-related matters is quite another.