Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - iruu

Pages: 1 2 [3] 4
31
BitShares PTS / Re: Windows Azure Free Trial
« on: November 14, 2013, 06:10:51 am »
How many pts/hour does this make? For how many hours?

A $200 free credit isn't free. It's going to be worth $200 if it turns out you need a cloud computing for a server or something. So as not to lose money, evaluating your probability of needing cloud and expected time of that event is needed.

32
General Discussion / Re: WARNING - Don't due any trades with MRodstein
« on: November 14, 2013, 12:27:23 am »
He responded in PM to my trade 150PTS@0.008, but went silent after I suggested only escrow/he pays first... so, seeing as he bought from you for 0.009...

33
Marketplace / [WTS] 150PTS@0.008, 50PTS minimum sale
« on: November 13, 2013, 04:31:38 pm »
Either you send first, or escrow (you pay the fee).

Please don't write PMs, write here.

I already made one successful trade (price rises though :) ) so you can trust me. I have many pts to sell.

http://bitsharestalk.org/index.php?topic=509.0

34
Hi. On a user page, could you add a "current unconfirmed balance" counter?
If you do, thank you.

Quote from: doesntmatter
really??? ypool 141593.60 cp/m right now... I notice there is no total for hashes or shares on your page...?
"spm": "3666.4"

Although, I think a simple "PTS/hour" would be better and clearer.

35
Very fast, excellent, would recommend

36
Marketplace / Re: [WTS] 100 PTS@0.0078, 0.78BTC total
« on: November 12, 2013, 09:00:09 pm »
You had me scared there  :)

37
Marketplace / Re: [WTS] 100 PTS@0.0078, 0.78BTC total
« on: November 12, 2013, 08:57:25 pm »
I just compared md5sum of protoshares-qt.exe with one from http://the-iland.net/static/downloads/protoshares-0.2.0.zip and they're identical

38
Marketplace / Re: [WTS] 100 PTS@0.0078, 0.78BTC total
« on: November 12, 2013, 08:49:08 pm »
3a06d06d67f3295d8829c70952540540353c9c77406e21b6be0698e9fb503e5d

wallet is v0.8.5.0 I think that's the last one?

39
Marketplace / Re: [WTS] 100 PTS@0.0078, 0.78BTC total
« on: November 12, 2013, 08:35:44 pm »
Sent

40
Marketplace / Re: [WTS] 100 PTS@0.0078, 0.78BTC total
« on: November 12, 2013, 08:20:32 pm »
My btc address is 14SozZa92vRGC6HmfqXVFDRR7vLUrNqvro

41
Marketplace / [WTS] 100 PTS@0.0078, 0.78BTC total
« on: November 12, 2013, 07:48:06 pm »
Either you send first, or escrow (you pay the fee).

Please don't write PMs, write here. 


42
It would be interesting to see a number of connections (workers) per generating address.

43
Not that I know the guy, but stealing the first deposit would be the dumbest scam ever, especially when such market is really in demand. The page looks nice. 
Probably just a bug.

44
General Discussion / Re: $5000 Bounty - Momentum Proof-of-Work Hacking
« on: November 10, 2013, 09:20:16 am »
So what are the precise terms of the $5000 bounty now? So that there're no doubts whether something meets it or not. 

Let's say a gpu miner uses x amount of memory and searches the 2^23 sha space in y seconds, on a modern graphics card. What're the x and y for the bounty? 
Or is memory unimportant, just that it can work effectively on a gpu? Effectively defined how?
Or is it the relation? When memory use is cut in half, speed drops less than a half (ie. relation is lower than linear). 

Can a proof of concept use simpler hash functions than sha512, if the algorithm is hash algorithm-agnostic? 

Or is it off?

45
General Discussion / Re: $5000 Bounty - Momentum Proof-of-Work Hacking
« on: November 08, 2013, 09:58:13 pm »
The cycle algorithm idea is wrong, there's no cycle in the 50 bit hash. The cycle is in 512 bit sha output, but that's equivalent to finding sha512 collisions.

Regardless of birthday algorithm, generating 2^23 sha512 on radeon 5870 (no oc) takes 180ms. 64 bit values and 1024 block makes parallelism on current compute units hard. That's close to output of new intel cpu. So gpus aren't going to win by a big margin, like with bitcoin's sha256. 
The collision finding time, even on a cpu, is almost insignificant, 2^26 is a small value. 
I'm writing gpu miner as an opencl learning experience. Looks like finding all birthdays in 2^26 hashes is going to take 200-400ms on radeon 5870.

edit: 70ms for sha512, so not that bad. (optimizing opencl is completely different from cpu, and I'm not talking about obvious sequential vs parallel differences), time for the harder part.

Pages: 1 2 [3] 4