Hey Christian,
Do you still have that FC wallet with the 6's or are they in an exchange already?
Do you still have that FC wallet with the 6's or are they in an exchange already?
I have the top 1 address and MOAR...
Topic: [ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX] - page 624. (Read 3426922 times)
March 31, 2014, 09:30:33 AM
I have the top 1 address and MOAR...
March 31, 2014, 09:27:06 AM
Mining is just going to keep moving from one algo to the next. Aslong as we can move quicker then tue asics can be built, they wont effect much too much. But because coins have such a long mining time it is. The only coins that will ever be asic resistant are those on an algo that asics arent for that are totally mined in less then a year. So asic devs dont have time to develop them
March 31, 2014, 09:23:47 AM
Damn you, warm spring weather! You might soon be the cause for me lowering the overclock on my GPU!
I've turned mine off to avert climate change. ;-)
Hey Christian,
Do you still have that FC wallet with the 6's or are they in an exchange already?
March 31, 2014, 09:21:43 AM
These chaps disgust me: https://litecointalk.org/index.php?topic=18166.0;topicseen
Clearly bribed by kncminer of whoever is supplying them with free shiny scrypt asics.
Clearly bribed by kncminer of whoever is supplying them with free shiny scrypt asics.
Both the devs have a good point though. Hark forking can have serious consequences and X11's longer confirmation times can be a problem. The scrypt miners have been at it for a month already. Check the difficulty raise this month. It's gone from 3500-6500 at it's peak. Almost doubled. This is just the beginning. More are coming and the first units that have been hashing from the beginning of this month will only be shipped late June/July. I,for one,can't imagine what's going to happen to mining at home when the ASICs are in full swing,say,by this September or October.
March 31, 2014, 09:01:12 AM
Damn you, warm spring weather! You might soon be the cause for me lowering the overclock on my GPU!
I've turned mine off to avert climate change. ;-)
March 31, 2014, 08:48:58 AM
Damn you, warm spring weather! You might soon be the cause for me lowering the overclock on my GPU!
March 31, 2014, 08:47:58 AM
These chaps disgust me: https://litecointalk.org/index.php?topic=18166.0;topicseen
Clearly bribed by kncminer of whoever is supplying them with free shiny scrypt asics.
Clearly bribed by kncminer of whoever is supplying them with free shiny scrypt asics.
March 31, 2014, 07:27:33 AM
March 31, 2014, 07:09:36 AM
A C-Coin (Christian-Coin), only mineable @ccminer@compute capability 5.0 (750Ti)would be cool.
and worthless.
March 31, 2014, 06:49:35 AM
Ever heard the word secret?
He wasn't impressed with the primegpu miner I mentioned, so it must be something big :p
sha256 at 1 THash with a gpu, that would be impressiveHe wasn't impressed with the primegpu miner I mentioned, so it must be something big :p
A C-Coin (Christian-Coin), only mineable @ccminer@compute capability 5.0 (750Ti)would be cool.
Christian
March 31, 2014, 06:27:49 AM
Ever heard the word secret?
He wasn't impressed with the primegpu miner I mentioned, so it must be something big :p
sha256 at 1 THash with a gpu, that would be impressive He wasn't impressed with the primegpu miner I mentioned, so it must be something big :p
March 31, 2014, 06:20:13 AM
Yep. We will get a random message saying some algo noone has ever heard of before(in the gpu mining realm) has been implemented and this thread will explode :p
March 31, 2014, 06:17:08 AM
Ever heard the word secret?
He wasn't impressed with the primegpu miner I mentioned, so it must be something big :p
He wasn't impressed with the primegpu miner I mentioned, so it must be something big :p
March 31, 2014, 06:04:37 AM
Such a tease. I want to know what the secret your working on is so badly. But im glad yiur keeping it hidden. If the amd guys figure out what your working on then they will try aswell
March 31, 2014, 06:02:54 AM
hey, any plans on making cudaminer x11 algorithm compatible? I've started mining Clear Water Coin today and there are talks of switching to x11 from scrypt, and most my rigs are running on cudaminer atm.. I never even heard of x11 until today.
x11 is a bit further down the roadmap for us...
Christian
March 31, 2014, 05:57:14 AM
Yacoin is where it's at 
I like that higher N factors tend to equalize hash rates between all miners, and make it solo mineable to a certain extent.
I like that higher N factors tend to equalize hash rates between all miners, and make it solo mineable to a certain extent.
I totally forgot about yac haha. It doesnt look too bad to mine from the profit graph above
March 31, 2014, 05:54:39 AM
Yacoin is where it's at
I like that higher N factors tend to equalize hash rates between all miners, and make it solo mineable to a certain extent.
I like that higher N factors tend to equalize hash rates between all miners, and make it solo mineable to a certain extent.
March 31, 2014, 04:34:59 AM
[
Surprising that all the "cuda coins" we did in ccMiner look pretty profitable in your charts.
I had almost written off fugueCoin already...
but beware, the buyer market for Fugue is weak, and it may drop sharply in price whenever
there's currently no buyer.
Remember to VOTE HERE https://www.cryptoaltex.com/index.php?page=newcoin for FugueCoin to succeed. The more exchanges the better this coin will do.
Also Yacoin is doing pretty good, despite the fact that the coin is currently sold under value (IMHO).
Christian
March 31, 2014, 03:39:52 AM
@cbuchner1
I have noticed something very strange between x86 and x64 of the 2-28 release regarding:
On a 750ti using x86 will result in pretty normal hashing.
On a 750ti using x64 results in probably 1/10th if not less the x86 performance.
As an add, I think cudaminer can make huge gains in performance if you can save all that bus bandwidth. But that's just my armchair coders opinion.
When I initially got my 750tis, I forgot to select pci 2.0 and I was running pci 3.0. While the machine could keep up, 5 of my 6 cards would do 375 to 400 khash on scrypt, the last card would do about 170. Eventually the system would eventually choke up due to an interrupt storm, but if those numbers are at all accurate I think cudaminer could squeeze out 25% more performance than we have now.
Again, I deffer to your expertise -- but I am very optimistic about future performance increases.
I have noticed something very strange between x86 and x64 of the 2-28 release regarding:
Quote
-a scrypt:2048 -l T5x20 -C 2 -b 4096 -i 0
On a 750ti using x86 will result in pretty normal hashing.
On a 750ti using x64 results in probably 1/10th if not less the x86 performance.
As an add, I think cudaminer can make huge gains in performance if you can save all that bus bandwidth. But that's just my armchair coders opinion.
When I initially got my 750tis, I forgot to select pci 2.0 and I was running pci 3.0. While the machine could keep up, 5 of my 6 cards would do 375 to 400 khash on scrypt, the last card would do about 170. Eventually the system would eventually choke up due to an interrupt storm, but if those numbers are at all accurate I think cudaminer could squeeze out 25% more performance than we have now.
Again, I deffer to your expertise -- but I am very optimistic about future performance increases.
March 31, 2014, 01:58:28 AM
I think we have to return to the good old Cudaminer...
@Christian
@Christian
Here's my tentative roadmap for continued development. Note the lack of a time table 
1) fix the issue with PCIe bandwidth, so people with 1x risers are more happy. CUDA can do all the
hash evaluation so we save a LOT of bus bandwidth.
2) Add blake256 for blakecoin and related coins.
3) look at implementing some failover options
4) see if I can get a monitoring API included
Any news, especially @1) ? 1) fix the issue with PCIe bandwidth, so people with 1x risers are more happy. CUDA can do all the
hash evaluation so we save a LOT of bus bandwidth.
2) Add blake256 for blakecoin and related coins.
3) look at implementing some failover options
4) see if I can get a monitoring API included
Jump to: