bitpar
Newbie
Offline
Activity: 59
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 12:09:05 AM |
|
how do we know he doesn't already have a GPU miner?
You don't. If it's easy to make one, then do so yourself. If it isn't, then either no one has one or someone spent a lot of time to make one and will therefore benefit.
|
|
|
|
tacotime
Legendary
Offline
Activity: 1484
Merit: 1005
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 12:15:29 AM |
|
See the original memcoin thread for why this is a bad idea https://bitcointalk.org/index.php?topic=122256.0
|
XMR: 44GBHzv6ZyQdJkjqZje6KLZ3xSyN1hBSFAnLP6EAqJtCRVzMzZmeXTC2AHKDS9aEDTRKmo6a6o9r9j86pYfhCWDkKjbtcns
|
|
|
chris200x9
Legendary
Offline
Activity: 1316
Merit: 1011
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 12:20:51 AM |
|
Looks interesting
|
|
|
|
alex_fun
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 12:23:04 AM |
|
so whats the exact release time in gmt say? May 06, 2013, 07:52:41 PM server time? ![Smiley](https://bitcointalk.org/Smileys/default/smiley.gif)
|
|
|
|
alex_fun
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 12:26:03 AM |
|
Also once released surely it wont take long fro gpu miners to twick miner? Anyone experimented with diff scrypt params here to see how cpu and gpu performs? ![Smiley](https://bitcointalk.org/Smileys/default/smiley.gif)
|
|
|
|
tacotime
Legendary
Offline
Activity: 1484
Merit: 1005
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 12:26:31 AM |
|
Another thing to note is that changing N changes the amount of memory required. This will (eventually) make gpu mining not worthwhile and much less efficient than cpu mining. ASICs and FPGAs (in their current renditions) won't even come close. This will be a coin ruled by fast memory, and lots of it.
Not really true. Space-time tradeoff will still allow efficient mining by fpga/asic/gpu.
|
XMR: 44GBHzv6ZyQdJkjqZje6KLZ3xSyN1hBSFAnLP6EAqJtCRVzMzZmeXTC2AHKDS9aEDTRKmo6a6o9r9j86pYfhCWDkKjbtcns
|
|
|
tacotime
Legendary
Offline
Activity: 1484
Merit: 1005
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 12:29:51 AM |
|
Also once released surely it wont take long fro gpu miners to twick miner? Anyone experimented with diff scrypt params here to see how cpu and gpu performs? ![Smiley](https://bitcointalk.org/Smileys/default/smiley.gif) As long as gpu memory has way more memory bandwidth than cpu, gpu will be much faster regardless of N value, I would expect. In fact, the greater the N value the greater a gpu will hash compared to cpu, because the dependence on memory bandwidth increases. There is data on hash time in the memcoin thread, once you hit the ddr3 memory with a cpu it's clear that the bandwidth of dram limits you.
|
XMR: 44GBHzv6ZyQdJkjqZje6KLZ3xSyN1hBSFAnLP6EAqJtCRVzMzZmeXTC2AHKDS9aEDTRKmo6a6o9r9j86pYfhCWDkKjbtcns
|
|
|
alex_fun
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 12:36:48 AM |
|
so what are those 3 scrypt params? gpu got ddr5 yes however not alot of it ![Smiley](https://bitcointalk.org/Smileys/default/smiley.gif) so that can help cpu computers with large ram as conventional ram is cheap ![Smiley](https://bitcointalk.org/Smileys/default/smiley.gif) Taco can you provide link to that thread you mention?
|
|
|
|
Joerii
Legendary
Offline
Activity: 1274
Merit: 1050
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 12:50:16 AM |
|
Not really true. Space-time tradeoff will still allow efficient mining by fpga/asic/gpu.
Due to the curvature of space-time, this solution would obviously create a singularity that will swallow earth and then the rest of the universe. ![Cheesy](https://bitcointalk.org/Smileys/default/cheesy.gif)
|
Hypercube - get the attention you deserve
|
|
|
tacotime
Legendary
Offline
Activity: 1484
Merit: 1005
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 01:34:25 AM |
|
so what are those 3 scrypt params? gpu got ddr5 yes however not alot of it ![Smiley](https://bitcointalk.org/Smileys/default/smiley.gif) so that can help cpu computers with large ram as conventional ram is cheap ![Smiley](https://bitcointalk.org/Smileys/default/smiley.gif) Taco can you provide link to that thread you mention? N = memory/computation complexity factor r = memory complexity factor p = parallelization factor (1 = hardest) Once you get into big memory sizes for CPU calculation of sCrypt hashes using either very large N or r values, you get get a massive dependence on memory bandwidth rather than any computational function. Due to the curvature of space-time, this solution would obviously create a singularity that will swallow earth and then the rest of the universe. ![Wink](https://bitcointalk.org/Smileys/default/wink.gif)
|
XMR: 44GBHzv6ZyQdJkjqZje6KLZ3xSyN1hBSFAnLP6EAqJtCRVzMzZmeXTC2AHKDS9aEDTRKmo6a6o9r9j86pYfhCWDkKjbtcns
|
|
|
alex_fun
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 02:08:54 AM |
|
thanks ![Smiley](https://bitcointalk.org/Smileys/default/smiley.gif)
|
|
|
|
Joerii
Legendary
Offline
Activity: 1274
Merit: 1050
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 02:35:18 AM |
|
So you're going through with this ? Will it be a test that you kill later or will it stay ? I'm not going to stop my rig just "for the lulz" you know.
|
Hypercube - get the attention you deserve
|
|
|
Koekiemonster
Sr. Member
![*](https://bitcointalk.org/Themes/custom1/images/star.gif) ![*](https://bitcointalk.org/Themes/custom1/images/star.gif) ![*](https://bitcointalk.org/Themes/custom1/images/star.gif)
Offline
Activity: 321
Merit: 250
Bitbuy.nl!
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 02:46:43 AM |
|
Watching
|
|
|
|
MyZhre
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 02:48:09 AM |
|
Interesting coin, let's see how its works
|
NXT: 13997163105778396158
|
|
|
juicingapple
Newbie
Offline
Activity: 14
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 02:52:15 AM |
|
dumb name, not appealing. Will fail.
|
|
|
|
meebs
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 03:22:40 AM |
|
Quick question, is there a CPU-mining program that can handle variable Scrypt rounds?
Edit: aside from the built-in miner, which is significantly slower.
if everyone is using the same software ot mine, the fact that it is slower will be irrelevant, as noone will have the advantage.
|
|
|
|
rbdrbd
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 03:41:49 AM |
|
Quick question, is there a CPU-mining program that can handle variable Scrypt rounds?
Edit: aside from the built-in miner, which is significantly slower.
if everyone is using the same software ot mine, the fact that it is slower will be irrelevant, as noone will have the advantage. Not if someone modifies the scrypt routines in cgminer to work similar to how it works in the built in yacoin miner. No contest. I give it a day at most before that happens with this coin.
|
|
|
|
Tobius
Member
![*](https://bitcointalk.org/Themes/custom1/images/star.gif)
Offline
Activity: 182
Merit: 10
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 04:41:42 AM |
|
dumb name, not appealing. Will fail.
As much as I hate to agree with you, He's right. Can you possibly consider changing the name to something more...professional? This could have some potential, but if the name is silly, it will die before the new year :\
|
|
|
|
Tobius
Member
![*](https://bitcointalk.org/Themes/custom1/images/star.gif)
Offline
Activity: 182
Merit: 10
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 04:42:35 AM |
|
Hell the biggest reasons I don't believe feathercoin and chinacoin don't have a future is that who the hell is going to use something called a feathercoin, or a chinacoin? It sounds childish
|
|
|
|
VelvetLeaf
Member
![*](https://bitcointalk.org/Themes/custom1/images/star.gif)
Offline
Activity: 98
Merit: 10
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 06, 2013, 04:44:43 AM |
|
Really someone just needs to implement it with high r values and see what kind of hash rates they get. I don't think an r leading to 512 MB of RAM required will be that catastrophic.
Here you go: [test]$ time ./scrypt-ref 1024 1 1 0.007u 0.000s 0:00.00 0.0% 0+0k 0+0io 0pf+0w [test]$ time ./scrypt-ref 1024 4096 1 31.282u 0.235s 0:31.55 99.8% 0+0k 0+0io 0pf+0w [test]$ time ./scrypt-sse 1024 4096 1 9.725u 0.225s 0:09.95 99.8% 0+0k 0+0io 0pf+0w [test]$ time ./scrypt-nosse 1024 4096 1 7.535u 0.210s 0:07.75 99.8% 0+0k 0+0io 0pf+0w [test]$ cat /proc/cpuinfo | grep name model name : Intel(R) Xeon(R) CPU E5540 @ 2.53GHz model name : Intel(R) Xeon(R) CPU E5540 @ 2.53GHz model name : Intel(R) Xeon(R) CPU E5540 @ 2.53GHz model name : Intel(R) Xeon(R) CPU E5540 @ 2.53GHz [test]$ uname -a Linux cs 2.6.18-308.16.1.el5 #1 SMP Tue Sep 18 07:21:07 EDT 2012 x86_64 x86_64 x86_64 GNU/Linux
About 8 seconds per hash attempt with 512 megabytes, so it'd take 18 days to verify the initial download of a blockchain with 200,000 blocks. ouch. Basically everyone would have to run it through an electrum server of some sort. I vote the name of the coin to be ElectrumCoin.
|
BTC : 1GN81dxzxyFPQsyAtdocXr5S9Mcg4wcfFG LTC : LgmYvXsYXc4xdjsMKXJWqtagxVvioK6iaw FC : 6dpSnKMtttUUYzaRu1EB7Lu18PBRVHU3V7
|
|
|
|