BTW, thanks for the tip tj

And Wooot, Catalyst 9.10 finally out ... trying them right away

Well, I don't think it would change something. It would definitely grow my "Vantage e-peen" but I don't care

In game, I would probably not notice any difference, probably only ~0.5FPS in selected games ... I think I'll wait for the 32nm i9 scheduled for Q1 next year for a mobo / CPU upgrade

---

Added on October 23, 2009, 6:06 pmCorrection: NFS SHIFT is still rubbish. It seems to depend of the track. Some I get a steady 60FPS (V-sync on) and some the FPS will dips below 20s and look like a slideshow

This post has been edited by r4st4m4n: Oct 23 2009, 06:07 PM

I've been thinking about that. But
1) I am not sure it will fix the problem, I need to some research to see if anyone is doing it.
2) Nvidia already try once to block it and although a workaround as been found this time, I am a bit worry that next time Nvidia manage to block it for good.
3) I am hoping that DX11 physics will send Nvidia PhysX to oblivion

Although I'd be interested to see if it works. Anyone who lend me a Nvidia card for testing ?

EDIT: me think I have a Quadro in my work PC which is actually a 8600GT, I wonder if I could use that for physx ... ?

---

Added on October 24, 2009, 7:48 pmI can't use a Quadro Card for Physx ... However I can fool the system by editing the INF and have the OS think I installed a 8600 GT instead of a Quadro. The problem now is that Vista doesn't allow to use 2 different VGA if they don't use the same drivers => not possible to have ATI and Nvidia working side by side. So now I have to install Win 7 if I want to try the hack ... :/

This post has been edited by r4st4m4n: Oct 24 2009, 07:48 PM

I've just finished installing 7

I'm installing the drivers and need also to test NFS before the PhysX hack ... will keep you updated

It seems the PhysX hack work as I am getting 1500pts more in Vantage due to the physics test being a lot better. However, the FPS still sucks in NFS

It doesn't seem to even run slightly better than before ... I guess I'll just have to wait for EA games to release a patch for ATI.

Thanks for the advice. Unfortunately I tried and it didn't seem to help. I just have to wait and hope EA release a patch ...


This post has been edited by r4st4m4n: Nov 10 2009, 09:28 PM

New update





Picture of the block installed coming soon as well as results

Yup, the compression fittings are next on my WCing shopping list, but I want to upgrade my monitor first.

Here are some pictures of the card in the case. The high flow fitting and metalic worm clamp give a kind of industrial look which is not bad either





I haven't had much time to test the OCing. But I did a quick testing @ 1.2V Vgpu and can push the clock to 1015 (gpu) / 1250 (GDDR5) and it seems stable as I could run Vantage and 2x3 loops of Crysis Warhead benchmark + ~1h30 of rfactor. I might be able to push even further at this voltage, But more important I still have plenty of room to push the voltage up to 1.35V


Win 7 - CCC 9.10 - HD5870@stock - Q6600@3.6GHz



Win 7 - CCC 9.10 - HD5870@1015-1250 - vgpu 1.2V - Q6600@3.6GHz



At 1015/1250 I've already got +1K points in Vantage P score, not bad at all

This post has been edited by r4st4m4n: Nov 11 2009, 04:03 PM

Same block ? Do you mind posting your config and some GPU-z screen while loading so I can compare your temp ?

9.12 improves NFS shift by 20% I got to download these drivers fast then. I have been waiting for some good NFS drivers for ages ...

This post has been edited by r4st4m4n: Nov 11 2009, 09:41 PM

Yup, me too it's bothering me these ugly wires. I have to get a dremel from reiben to cut a hole in the middle plate to hide all these wires. I think what is putting me off is just that I need to take the whole case apart to do that and I'm kind of lazy / don't have too much time on my hands at the moment. But it will come

It's an EK (FC5870, you can see the pictures of the block few posts earlier) and It doesn't come with any fitting, only 2 plugs. I am using D-tek high flow fitting ATM.

I would stay away from bitspower. Their design for the 5850/5870 block sucks:




And the temps seems to be awful compared to the EK or Swiftech block:



You can read the posts regarding bitspower block starting from there: http://www.xtremesystems.org/forums/showpo...7&postcount=395

That's only one user's opinion though, so it may not be true for everybody, but still the internal of the block looks poorly designed ...

You can check how performs the other block few pages earlier in the same thread.

This post has been edited by r4st4m4n: Nov 14 2009, 02:22 PM

Yup, both are right. If you buy tubing made for WCing, it is much easier to route in your case and their diameter matches the standards used in WCing. They are also tested to make sure that they do not react with the liquid and therefore do not cause any chemical reaction with your block (corrosion, deposit, ...)l.

And frankly, the tube is the cheapest component in the loop. You can find Duralene tubing for RM 10 / foot and you generally needs not more than 8 feet for the loop depending of the case and how your route the tubing. So it is always better to pay a little extra compared to classic gardening hose, and to be sure your tubing won't affect the others expensive components like the blocks and pump.

Oh yeah, forgot to post new pix about that.

When I drained my loop for the EK block, I realized that I started to get AGAIN algae in my loop. It was only about 3 weeks since I cleaned my loop and this time I used the anti-algae that bryanyeo87 is selling (DraconisThermoX thinkgy) ... :/

So I changed again the tubing and this time I've put like 5 times the recommended usage (I think around 20+ drops in 1.5L vs the 3-4 drops in 1L recommended ). I am still using the spritzer distilled water but bought new 1.5L bottle. I'm not sure if it mattered but last time I used a 5L bottle (non opened) that I had for few months. Plus this time I made sure to bleed the loop correctly by shaking the case. Last time I didn't shake the case and I had some pockets of air coming out after 1-2 weeks so I had to open the res to top-up. This time I only had to top up the loop once after running the loop few hours and I should no have to open the res again to top up. I hope it will prevent any contamination.

So if my loop get contaminated again, I guess I'll buy some well-known anti-algae product like PT Nuke but I really hope that this time it will be okay ...

Agreed it does look nice
I personally go for performance first and look in second, that's why I chose EKWB. But you may have other priority

To be fair, the temperature are still better than the stock cooler, just not what we could expect from a WB.

EDIT: hmm wait, is he using a water chiller for the VGA ? From the picture, we can see a COOL IT window with a "chiller A" temp. If the water is chilled for the VGA too, then we can say the temperature are really bad ...

This post has been edited by r4st4m4n: Nov 15 2009, 10:05 PM

I can't say as I haven't installed the MCW-60 on the HD 5870. I wanted too, but realized that it would be difficult to stick sinks on the smaller chip on the card. I didn't want to take any risk of overheating so I have waited to receive the full cover WB before to WCed the card.

You can probably expect better temp because you are not getting the heat from the other components, and also the MCW-60 ha better design for cooling (pin matrix) at the expense of more restriction. But these card do require good cooling on the mosfets, particularly when overclocking, and a few degrees less on the GPU generally don't get you a big difference in clocks. So I think overall the full cover WB is a better option. It also avoid the need for a fan blowing on the memory/mosfets.

This post has been edited by r4st4m4n: Nov 15 2009, 10:21 PM

=> Water chiller. He says in his post "[...]and my water stays at 20C [...]" + in his signature he has "Dual Freezone Elite" which are water chiller. So when taking into consideration that the water is at 20*C, the block is performing pretty badly. Meaning in Malaysia with a classic WC system, the block might perform just on par with the stock cooler ... :/

This post has been edited by r4st4m4n: Nov 16 2009, 12:07 AM

Nope, no major update since the water block. I am running happily at 1010/1250 (GPU/mem) with the GPU@1.2V. I think it can run 1020~1025 stable at this voltage but I backed up few Mhz for 24/7 usage. At 1010, It is rock stable even after hours of Crysis/NFS Shift/Dirt 2

I've actually wanted to see how far I could go but didn't have much time for that. And 1+Ghz @ 1.2V is probably the sweet spot as the very quick testing I did showed little clock improvement with more volts.

Thanks for the advice, i didn't know that.

My tube have again become a bit cloudy, but not as bad as last time. Now I am starting to wonder if it is really algae since my res stays completely clear. So I'm thinking it might be some residue of flux from the rad. Because the tube "absorbs" stuff unlike the res, that might be why I see it on the tube but not on the res. I'm going to wait a little while to see if it settles like that and I'm probably going to switch to white tubing ...


Yup, I saw the "LT+" trick on xtremesystems. But I'm quite happy with my temp so I don't think it is worth it because of the added restriction which might defeat the gain from the mod.

Thanks

I'd say go for it. The TJ-07 is definitively worth the money.

My NFS shift issue has been fixed with the 1.2 patch already. And I just realized yesterday the 9.12 were out so I haven't really tested them yet