I found a new version of OpenCL (AMD APP, Stream SDK, whatever they want to call it) in a leaked version of a beta driver. This driver supposedly gives Windows 8 support. Anyway, I don't give a bleep about the driver, just the OpenCL versions. On Catalyst 12.3 with OpenCL version "2.6" my DLL versions are: amdocl(64).dll 10.0.831.4 opencl.dll 1.2.0.0 From this leaked driver the DLL versions are: amdocl(64).dll 10.0.898.1 opencl.dll 1.2.1.0 Uninstalled version "2.6", deleted the DLL's manually in System32 and SysWOW64 (necessary step going between versions!), then installed this new leaked version. Deleted my kernel bins in cgminer dir as usual, then launched... Can't really see any change in performance. Hopefully this gives ckolivas and the kernel devs something to chew on and progress forward, though. Driver link: http://jumbofiles.com/h2l4qczhjkb4I will host just the 32 and 64bit versions of the OpenCL.msi myself for those who don't care about the whole driver: http://www.delta9IT.com/OpenCL32.rar http://www.delta9IT.com/OpenCL64.rar |
|
|
|
Cool, bro. You can argue semantics and provide poor analogies.
How about you just explain how 'two window fans' provide an adequate volume of air to compensate for the delta-t of 40F. You do this for a living; I'd love to see the calculations.
It's a delta-t of 65.2F. I don't run my air-cooled GPUs at 60C. I'd like the fans to last more than a few months and I don't need the GPUs to keep running for decades. I'm not doing your homework. Judging by your previous posts, you'd turn mine into a conspiracy theory. If you knew enough about the subject to understand the calculations one would post, you'd realize that a temperature differential of 65F isn't really THAT difficult to work with when you have a decent supply of fresh air. I'm not showing you my birth certificate either. I grow dope and fish for a living, bro. I have no idea what volume of cross-flow you would need to dissipate 21.5kW of heat with 100F ambient temperatures, nor how to calculate it. I can say that I have dealt with ~25kw of indoor lighting, and it took a six-ton compressor working it's ass off and fans that no one on earth would call 'window fans' even when the intake temps were in the sixties(F). Cannabis and fish together or separately? I'm trying to get my aquaponics setup going right now, but it will just grow leafy greens and maybe peppers. Hard to find edible fish to grow around here too, so I just have some hobby style fish. I have set up a cannabis aquaponics system, but it's too finicky, and requires too much space. Peppers get tall and kinda bulky. You will get a lot more productivity sticking with herbs and greens. Where do you live that you can't get edible fish? Tilapia are pretty easy to snag except in Florida, and they are completely fucking indestructible. I want to grow tilapia but they require a license here in California. I think it's cool you actually did a cannabis aquaponics system tho! |
|
|
|
|
I ran a couple numbers. You end up paying more for the card leasing it than retail in 12 months, obviously. The price per BTC needs to go up a little and stay above that level to make the lease vs profit ratio positive. If BTC were $10 it would be a no brainer /wish
|
|
|
|
|
The card doesn't have to get the power through the slot, it will get it through the dedicated power connectors. If you really want to you can buy the cable from cablesaurus that injects power using a molex connector into the pci express extension cable.
|
|
|
|
Lease to own or upgrade video cards whenever you want. (Anywhere) Date: 2012-03-01, 4:02PM PST Reply to: jzvpd-2879763486@sale.craigslist.org [Errors when replying to ads?] Do you want the newest video card from AMD or Nvidia, but just can't afford to shell out $500-$800 for a new card? I have a solution. I am in no way affiliated with this company, but I do lease from them and it's excellent. For example, I just leased a Saphire 7970 for around $60 a month for 12 months, but check this out, the 7990 is supposed to come out this month, well no worries, as soon as it comes out I can just ship back the 7970, (which they pay for), and pay the monthly difference and they send me the 7990 at no additional charge. It really is a great service, no BS. Now, here's what I get out of this. For every person I refer to them, I get $5 credited to my account, but you get rewarded too, you also get $5 for mentioning my name. Email me for details if interested, this is real. Oh, they do take debit cards, you must be 18 or older. Location: Anywhere it's NOT ok to contact this poster with services or other commercial interests PostingID: 2879763486 OK, so someone lease a ton of 7970's and tell me how much you make! lol |
|
|
|
Ok so I shouldn't tell you how I brought my personal video card to work and used it to mine bitcoins while my new video cards mined at home. Got it! Got busted for that one too, this time I did not try to hide anything more for the IT boss/director  |
|
|
|
|
I remember the old days when I used to run Folding @ Home on 100+ computers at work. The goal was to seriously find a cure for cancer. There was no financial gain back then. Boss found out after a while that every computer had 100% CPU usage, but no process listed, lol. I explained to him the project, how I did it all automatically with scripts, etc. I had to take down the cluster, but my boss was seriously impressed with my skills on how to handle a big cluster of folding machines. He later used my scripts that I developed to push out certain new software. It was such a win situation! I feel bad for these IT guys who get busted for bitcoin mining and the bosses don't recognize they actually have skills that can be guided towards being more productive. What ever happened to hackers that got hired by the company they hacked into because they proved their skills!?!?
|
|
|
|
OK, you found the other brand that is a clone of the Iwaki. I knew from my saltwater aquarium days there were 2 Iwaki clones, one was Pan World and the other I couldn't think of for the life of me so I didn't list it. I would consider a BlueLine a contender also. They all share excellent the same engineering specs, with different people actually making them. Fake edit: I see you bought a used PanWorld 150. Badass! As long as the price isn't outrageous it is still cool that you got an overkill pump. And yes, rjk, VFD would be the ultimate luxury on a setup like this. Totally unnecessary, but we are nerds, so every improvement is cool! Should put one of these in the system: http://www.marinedepot.com/Digital_Aquatics_ReefKeeper_Lite_Controller_Multi_Item_Monitors_Controllers_for_Saltwater_Aquariums-Digital_Aquatics-DA1131-FITEMOMI-vi.html . I mean, wouldn't it be cool to monitor the pH, temp, and *salinity* of your *freshwater* cooling loop? Im just kidding man, seriously, just kidding. I want to see this project grow. |
|
|
|
Wasn't the claim that one could cool 20KW of heat load with a pair of window fans in 100F ambient temps? Funny how the topic has changed to custom building design, 3 square foot holes in the roof, 2000 watt blowers, datacenters in sweden, "anything can be cooled with enough airflow", how houses get cold when you open the window in winter, and other distractions. What happened to 20KW heat load, pair of window fans and 100F ambient temps?  Best summary of what happened, IMO. |
|
|
|
Cool, bro. You can argue semantics and provide poor analogies.
How about you just explain how 'two window fans' provide an adequate volume of air to compensate for the delta-t of 40F. You do this for a living; I'd love to see the calculations.
It's a delta-t of 65.2F. I don't run my air-cooled GPUs at 60C. I'd like the fans to last more than a few months and I don't need the GPUs to keep running for decades. I'm not doing your homework. Judging by your previous posts, you'd turn mine into a conspiracy theory. If you knew enough about the subject to understand the calculations one would post, you'd realize that a temperature differential of 65F isn't really THAT difficult to work with when you have a decent supply of fresh air. I'm not showing you my birth certificate either. I grow dope and fish for a living, bro. I have no idea what volume of cross-flow you would need to dissipate 21.5kW of heat with 100F ambient temperatures, nor how to calculate it. I can say that I have dealt with ~25kw of indoor lighting, and it took a six-ton compressor working it's ass off and fans that no one on earth would call 'window fans' even when the intake temps were in the sixties(F). Cannabis and fish together or separately? I'm trying to get my aquaponics setup going right now, but it will just grow leafy greens and maybe peppers. Hard to find edible fish to grow around here too, so I just have some hobby style fish. |
|
|
|
Are you people
A - Jealous
B - Assholes to everyone you meet in "real life" too
or
C - Just completely ignorant?
B - Because when it comes to topics like this, I have way more experience than anyone I meet in a typical day. Now when I talk to people who have big titles in the enterprise computing world, *that's* when I start to listen instead of school. |
|
|
|
That article was more theory than proven practice. They only had 5 servers...in a tent, using way less power than something loaded with GPU's. Lets try to scale that up and see what happens.
[...]
And yes, it just so happens that I have had to design data centers in my career. Including power and cooling for them!
Then, as a data center designer, could you kindly explain to me what, exactly, fails to "scale up" when using outside ambient air to cool? Just how high of a server density do you need before "outside" experiences a significant increase in temperature? Trust me, Google and Facebook have looked very hard and written great articles about it.
Yes, they have. And you have apparently read them, because you specifically mention their impressive improvements on the boring ol' "swamp cooler". So no doubt, you know all about Facebook's newest DC in Lulea, where they expect to need less than two weeks of supplemental active cooling per year. Now compare the form factor and uptime demands of a hardcore miner against a Facebook datacenter - And try to tell me with a straight face that you don't see just the teensiest difference between a room full of mid-tower PCs loaded with GPUs that can go down for an hour or two in mid-afternoon on the hottest days of the year with no real harm done, vs row after row after freakin' row of 24core x 42U racks with a contractually guaranteed six-nines uptime? The difference I see is that we are packing much more power draw (read: heat generation) into a smaller space than what has been pointed out in this thread. Now you are comparing to 24 cores in a single chassis, but lets be more realistic and reference the MSDN tent article again. Modern servers are awesome with how much CPU they can cram into a 2u/4u form factor, or whatever servers you have, but they are also extremely efficient on power use nowadays. Also, the load on a typical server is very random and usually nowhere near 100%, unless you are running simulations or rendering Pixar movies or whatever. We load to 100% 24/7 with fire breathing GPU's packed right next to each other. What runs stable in that little tent they used cannot be compared to racks of GPU's the way we do it. Pulling air in a window on one side of a house and exhausting it through a window on the other side where you mining operation is just cannot be done after a certain level of power draw/heat generation. |
|
|
|
|
DeathAndTaxes, I would like to see sveetsnelda cool a large, dense mining cluster with 100F intake air. All he needs is enough airflow, right? lol. Have fun with the power and noise required to move that much air. This is where water steps in and becomes way more efficient. You still have to use forced airflow to dump the heat in the end, but it is NOTHING like designing a large scale air only system. I'm glad he designs data centers for a living also. He should tell Facebook they did it all wrong with their latest DC.
Edit: I missed where he said on the last page in this thread where he lives in a dry climate that gets to 100F and he had to use evap to control the temps. lol
|
|
|
|
|
A 7kW heat exchanger doesn't need to be huge. My go kart produces more heat and the radiator is tiny and doesn't even have a fan =P jk, bad example. Just FYI, a trick I learned on radiators in general is that the thicker they are the harder it is to push air through them to make them efficient. You can have a heat exchanger that dumps 20kW and is only 12 sq. in., but is so thick you need a serious freakin' fan just to push enough air through it (or 100mph of wind hehe). In your application you have all the space in the world to spread out the surface area, so a heat exchanger that is only 1/2" thick or whatever would be way easier to deal with.
|
|
|
|
I know it's too late, but you spent way too much on a pump. Iwaki is premium shit in the aquarium world. There is a clone available for each Iwaki model made by Pan World. Pan World was started by a former chief engineer from Iwaki. They just make the same pumps essentially, at way better prices. Also, I really doubt you had to go to a full external style pump. Probably could have gotten away with a beefy submersed pump and put it in a sump, for like half the money. The goal in the end is ROI, right?
Not to late. Haven't pulled the trigger on the pump. Was doing work on the heat exchanger and sweating npt connections this weekend. Show me a pump with a) similar curve as a Iwaki MD-40 b) has 1" NPT intake and discharge c) has similar heat dump d) has significantly lower price and I am interested. I have looked any haven't found it. Everything I have found either has horrible head (most aquarium pumps) or is horribly inefficient (most sump dumps) dropping 300W to 500W of heat into the loop or is more expensive than the Iwaki. If you found a good pump that can get similar flow as the Iwaki MD-40 at 10ft to 15ft of head let me know. Oh cool, you are still deciding on the pump. I'm going to be fairly blunt about this. What makes you think you need to pump ~15ft of head at 450+ GPH? Have you calculated your total head based on real height plus all your fittings and doo-dads? There are calculators out there, btw. Based on my experience with even shitty aquarium pumps, they are all underrated on what they will flow at their "supposed" maximum head height. I think you would be surprised what a ~50-75W submersed pump can do even "choked" down to 1/2" tubing. Don't get caught up too much in the system designs for water cooling in the overclocking computer world. They tend to grossly oversize EVERYTHING in search of dropping that last 1 degree. I was doing computer watercooling before you could just go out and buy a kit in a box. Some of the setups the enthusiasts started making in the early 2000's up to now are downright hilarious on how oversized everything is. Some of the more commercial application are more realistic, with way smaller tubing, pumps, and radiators. I think if you shoot for 50-60C temps at the GPU itself, instead of 2C above ambient like some guys, you would be surprised how cheap your initial investment gets on the equipment, host fast you get your ROI, and how long everything would last. If you live in a dry environment you could even look into evap cooling to dump the heat instead of a radiator. You know, how nuclear power plants are cooled? A while back people designed PVC "water bongs" to blow air up over water that was showering down into a sump. I made what I called a "bucket bong" that did the same thing in a 5 gallon bucket, just a shower head and an 80mm fan. The advantage? Your water coming out of the "bong" was 1-2C *colder* than ambient. The disadvantage? You have a bit of humidity to worry about in the room you dump the heat in. You could either plumb it outside, or just dump some of the air outside with another window fan again, lol. |
|
|
|
A proper ball bearing fan should last a really long time. It's the cheapo sleeve bearing fans placed on GPU's that fail when pushed hard.
Then I should consider myself lucky, they are "double ball bearings", a lot of balls. I love that marketing gimmick on small fans, lol. They act like it's an industrial HVAC fan with a foot long shaft that needs to be supported on both ends or something. Might as well say something like "DOUBLE BALL BEARING TECHNOLOGY! -- Our 120mm fans have a huge shaft that requires a lot of balls" |
|
|
|
|
I'll be honest I haven't made money trading lately like I used to because the price is so stable. I have been looking at adding hashing power to my rig for the last 8 hours. I thought FPGA's might have hit a sweet spot in ROI, but I'm back to looking at 7970's. I think I should just grow medical cannabis instead.
|
|
|
|
|
That article was more theory than proven practice. They only had 5 servers...in a tent, using way less power than something loaded with GPU's. Lets try to scale that up and see what happens.
There are *very* few places on earth where you can get away with cooling a dense DC with just ambient outside air. Trust me, Google and Facebook have looked very hard and written great articles about it. In the end they are forced to put DC's in places where the local temps get too hot, within the course of a year, to work with just outside air. We aren't talking about running a bunch of mining rigs just in the dead of winter. And when I say too hot it's not like I'm talking 135F. They end up using some form of energy input into the system to get the temps down. You would be surprised what Facebook can do by misting water into their fresh/recirc air. But that still takes power to pump! Not to mention the cost of the water!
And yes, it just so happens that I have had to design data centers in my career. Including power and cooling for them!
|
|
|
|
|
Show Posts
