Joke is on-me

[nss000]

Gents:

After all the amd-vs-INTEL CPU discussions I finally decided on an X-965/MSI_G70 combo at NewEGG. Silly me. About $390 not a bad price.

After a merry-go-round ordering attempt I chatted with a NewEGG rep who assured me their site was quite incompatible with FFox/Linux. Oh yeah ? I then offered to switch to OPERA if better ... and the NE guy said he'd never heard of it! Why don't I try Internet Explorer ...

I don't know whether to laugh or cry ................ but the money stays in my wallet. Likewise that for an nv_9800, Hanns 28" monitor, 650W power supply and couple-a 750G hard-drives. Perhaps the computer-gawds of THW were looking out for me.

 

[dunklegend]

Gents:

After all the amd-vs-INTEL CPU discussions I finally decided on an X-965/MSI_G70 combo at NewEGG. Silly me. About $390 not a bad price.

After a merry-go-round ordering attempt I chatted with a NewEGG rep who assured me their site was quite incompatible with FFox/Linux. Oh yeah ? I then offered to switch to OPERA if better ... and the NE guy said he'd never heard of it! Why don't I try Internet Explorer ...

I don't know whether to laugh or cry ................ but the money stays in my wallet. Likewise that for an nv_9800, Hanns 28" monitor, 650W power supply and couple-a 750G hard-drives. Perhaps the computer-gawds of THW were looking out for me.

I have always ordered from newegg in firefox, I don't know why did you have any problems.

 

[dunklegend]

Your spending 390 and not getting an I7? Should just burn the money instead

True, true

 

[nss000]

I haven't the foggest idea nor do I care. It's really not-in-my-job description to make NeweGGs online system work. They don't do me a favor by selling to me tho that's implied by their arrogance.

Anyrate as I suggested perhaps THW-computer.gawds were protecting me from a rash choice. Whichever ... I do know that $1000-bill stays in my wallet awaiting a more genial buying environment.

 

[nss000]

Your spending 390 and not getting an I7? Should just burn the money instead

I will take your criticism under advisement. An i7 really? That means I must also buy & install an aftermarket Zalman or Xigmatic. Gulp!

The comments are appreciated. As mentioned previously we ( modestly skilled ) "infrequent" builders truly COUNT-ON
the experience of more frequent, more aggressive kit experimenters. You-guys build 5-times to every one of ours. We are not "untrainable", but an element of luck does influence our success.

Perhaps my NewEgg order-failure is one-such lucky moment. Again thanks to all for sharp-eyed critique.

 

[warmon6]

I will take your criticism under advisement. An i7 really? That means I must also buy & install an aftermarket Zalman or Xigmatic. Gulp!

The comments are appreciated. As mentioned previously we ( modestly skilled ) "infrequent" builders truly COUNT-ON
the experience of more frequent, more aggressive kit experimenters. You-guys build 5-times to every one of ours. We are not "untrainable", but an element of luck does influence our success.

Perhaps my NewEgg order-failure is one-such lucky moment. Again thanks to all for sharp-eyed critique.

Unless you plan on doing major over clocking, then stock cooler will be fine for quite a while.

 

[choujij]

I put together a tower with a Core I7 920, on a high-end Gigabyte board for a friend. I also threw in a Phenom II X4 965 BE on a $100 Asus board in my HP tower for myself.

No over-clocking on either, we ran Everest on both.

-The I7 failed at roughly 40% into the stability test, (the temperatures reached around 62-63c per core). This surprised me since it was in a nice aftermarket tower with many fans for dispersing of ambient heat.

-The X4 passed running at 52c steady per core, peaking at 53c ever few seconds. This surprised me as well, since I figured in this HP case the sub-par cooling would give the X4 a disadvantage.


While initially the thought about pitting up an over-clocked chip against one at stock speeds seems unfair, the reality is (or at least in my experience) that if the faster chip cannot maintain stability using stock hardware and speeds, the test of stock speeds is still unfair.

I now believe that if I turn up my clock speed on my X4 to the point where it will fail a stability test very similar to how the I7 failed, sure it will out perform the I7. While I can't speak for everyone, in my experience for an "out of the box" product, the AMD can easily be overclocked where as the I7 requires much better cooling even at stock levels.

 

[cjl]

The i7 at 62-63C should be completely fine. That is nowhere near its thermal limit. I'm not sure why it failed, but it certainly wasn't temperature related.

 

[choujij]

One of the reasons I believe the cooling was the culprit, is because the heat sink that came with is I7 is terrible. The corners of the I7 are exposed, where as the heat sink that came with my X4 completely covers the chip.

 

[cjl]

That really says nothing about the quality of the heatsink, honestly. The heat is generated on the die itself, which is at the center of the heatspreader, and those temperatures are fine. The i7 heatsink isn't as good as the x4 heatsink, but it is definitely adequate for stock speed.

 

[choujij]

That really says nothing about the quality of the heatsink, honestly. The heat is generated on the die itself, which is at the center of the heatspreader, and those temperatures are fine. The i7 heatsink isn't as good as the x4 heatsink, but it is definitely adequate for stock speed.

Don't get me wrong, I'm not trying to knock the I7, but the only other thing I could blame would be the GigaByte board.


Perhaps you have some suggestions as to why it failed? FYI, we were only testing the cpu stability...

 

[jonsy2k]

Could be possible you got a bad chip, temp sensors are off, heatsink not mounted correctly, its hard to tell. Have you run the test several times to confirm that it fails at the same temperature always? You could also try other stress testing software, prime95 etc.

There is really no reason to upgrade from the stock cooler if your not going to overclock. As cjl said before, the heat is generated from the center of the chip anyway, and neither is 62-64 that high.

 

[nss000]

True, true

When I do-the-numbers: i7-920: $280+
Zalman: $50 < hot.si cpu >
ASROCK MB $200 < bulletproof 5-EGGS >

That's $530 INTEL kernel-kit. Where is that $390 you speak of ? Perhaps for a kit_wizard, but not me.

 

[jonsy2k]

When I do-the-numbers: i7-920: $280+
Zalman: $50 < hot.si cpu >
ASROCK MB $200 < bulletproof 5-EGGS >

That's $530 INTEL kernel-kit. Where is that $390 you speak of ? Perhaps for a kit_wizard, but not me.

So is there any reason you chose to ignore all of the posts that stated that you don't need an aftermarket cooler as long as your not doing extensive overclocking?

Your right, even without the cooler, the 1366 platform just costs more, but your paying for much improved performance over the 955.

 

[nss000]

So is there any reason you chose to ignore all of the posts that stated that you don't need an aftermarket cooler as long as your not doing extensive overclocking?

Your right, even without the cooler, the 1366 platform just costs more, but your paying for much improved performance over the 955.

Yes I appreciate your point. I have not ignored them ... I have taken notice of MANY usr_posts on various boards and reviews noting that i7-920s run-hot. And the retail heat_sink makes risk_adverse builders uncomfortable.
It's not a syllogism, but it is a threat. If I pay extra for it, then likely at some time I'll overclock and it will run hotter! A later HS/fan replacement could well be a horror ... I've read stories... it's not an option for me.

You may say so ... and INTEL may act so ... but as I observe it WEB "group wisdom" calls-to-question the retail-cooler from INTEL. Under any use it's hot. Hot = risk. Risk = threat. Why should I "build-in" a threat? If I can't afford to "buy-out" of that (heat)threat, then I look for less_risky kit. Pardon my "fuzzy logic".

Yep -- as best as I can judge it I'll buy high value, low risk kit that works the-first-time now ... and will be cranking away five years from now w/o significant additions.

 

[nss000]

Dont need that board. Replace it with this one

http://www.newegg.com/Product/Product.aspx?Item=N82E16813130227

I love how many people bitch about how much more the I7 costs. It's the best you can get today. ofcourse its going to be more

The (flawless) ASROCK lists for the same price < $170 > on NewEgg ... I quoted another E-tailer price as I currently cannot access NEWEGG purchase system ... cause unknown.

For me a couple-hundred bucks really is a big deal.

 

[brandonvi]

nss000 realy anyone that is telling you you cant run the thing with stock at normal speeds is an idiot i have seen tests that show the stock VS aftermarket heatsinks there was only like a 2-3C if not less at stock speeds it only really helped at all when the speeds where higher like a I7 920 turned up to 3.2 or 3.8Ghz

 

[choujij]

Could be possible you got a bad chip, temp sensors are off, heatsink not mounted correctly, its hard to tell. Have you run the test several times to confirm that it fails at the same temperature always? You could also try other stress testing software, prime95 etc.

I used Artic Silver 5, and triple checked my work. While I know that the heat sink is mounted correctly, I do not like the unorthodox way it is done (pressing down on the tabs). I say this because it felt as if something could easily break or go wrong.

Regarding the memory being used, it's Mushkin ram. It was used because my friend swears by it.

The temp sensors could be off or be misread by Everest, however that would still imply either the stock speed is a bit high or this particular chip is defective.

I will ask my friend to perform the tests again using another program. We will check the ram as well.


Thanks.

 

[brandonvi]

more then likly that chip is defective really at stock speed the thing should never no mater what you do even come close to over heat

well that is if as you say you have done it right