Please post here anything else (not relating to Maxwell technical matters)
User avatar
By b-kandor
#234365
Things seem very very expensive in europe - then doubled again in england!
User avatar
By b-kandor
#234808
My quad core is up and running - one quick question for comparison:

Using core-temp I'm seeing

55c
58c
52c
53c

When rendering in maxwell. I'm surprised to see so much variation between cores - is this common?

Thanks,

Kandor
By numerobis
#234814
yes this is normal. the quads have two chips under the heatspreader (here 1. 55+58° and 2. 52+53°) and it depends on how well the heatspreader fits on both of them.
the 58 to 52° difference is not optimal, but it should be no problem. but the 58° core could be the weak point if you plan to overclock...
i have around 4° difference between the lowest and highest temp.
to preclude a bad mounted heatsink you could reseat it (and maybe apply a good thermal compound - e.g. Arctic silver 5 - if you haven't already done.)

but your temps look good - assuming that you are using air cooling, and considering that it is summer. so it could be that remounting will have no effect...
and 58° should be no problem for a quad - from 70° on you can start to worry
i can easily reach 60-68° with my 3ghz q6600 - depends on the ambient temp.

...so, and now look how far you can go :mrgreen:
User avatar
By b-kandor
#234818
Thanks for the info!

I will overclock once it has a chance to set. I'm used to the e6400 chip that I had overclocked 50% to 3.2ghz (from 2.1) - it use to render around 52c per core (overclocked) and in the mid 40's without the clock.

Here I'm already hotter at stock.

I have used AS5 for years but on this system I tried the zalman stg1 grease which everyone is saying is as good or better than AS5. I'm using a cnsp9700 on my quadcore chip. It's the first day so once the compound sets (in a week) I may see better temps.

Ps - it's around 34c here today (record in canada) - but my office is thankfully in the lower level of my house with air conditioning 8)

Pps - how are you oc'ing yours? Are you changing the multiplier or just the freq.?
By jespi
#234824
Hello guys! Sorry for my novice question but it is possible install two quad 6600 together(8 cores), i mean, i´ve been looking for mother boards that support 2 quad 6600 but i can not find it.

Thanks in advance.
By numerobis
#234827
yes, the quads run very hot... :wink:

i have increased the fsb - higher multiplier only works for the qX's... normally.
...and higher fsb = faster system :mrgreen:

Tcase max for q66 is 62.2°C (ihs temp)
http://processorfinder.intel.com/detail ... Spec=SL9UM

so, i think ~15° more for the coretemp should be ok, but i don't it know exactly
By numerobis
#234829
jespi wrote:Hello guys! Sorry for my novice question but it is possible install two quad 6600 together(8 cores), i mean, i´ve been looking for mother boards that support 2 quad 6600 but i can not find it.

Thanks in advance.
no, it's only possible with the expensive 53xx Xeons for socket 771... :?
and you need registered ram for these boards
By jespi
#234836
:? Thanks numerobis, them i´ll buy a 6600 quad when the price low.
User avatar
By b-kandor
#234838
Rendering in modo at stock (modo is hotter than maxwell) I get

49c in speedfan and then speedfan lists 4 core temps that seem too low:

41c
45c
40c
40c

Meanwhile coretemp reads:

56c
60c
55c
55c

So not a lot of headroom there - I'll let the compound settle for a few days.
By numerobis
#234851
the speedfan temp should be Tcase or cover temp, coretemp reads like the name says the coretemp :D
coretemp should be more exact - speedfan only reads the bios temp

so - theoretically - with ~40° Tc there is ~20° headroom up to 62° :roll:
User avatar
By b-kandor
#234866
I thought the speedfan temp ~49C would be the Tcase so hence I'm thinking 10-12 headroom - which is still fine. 8)

ok thanks for explaining. actually I do copy the T[…]

Sketchup 2026 Released

Fernando wrote: " Now that Maxwell for Cinema[…]

Hello Gaspare, I could test the plugin on Rhino 8[…]

Hello Blanchett, I could reproduce the problem he[…]