mamad
Newbie
Offline
Activity: 17
Merit: 0
|
|
May 25, 2011, 06:12:20 PM |
|
First things first, I really love the idea. But I want to know if anyone is having any success with this. I just started running this on one site that has about 200 online users. What can I expect in terms of returns? it feels like the hash rates are so small that it wouldn't make a difference even with thousands of users online.
Thanks, Mamad
|
|
|
|
Turix
|
|
May 25, 2011, 06:14:28 PM |
|
First things first, I really love the idea. But I want to know if anyone is having any success with this. I just started running this on one site that has about 200 online users. What can I expect in terms of returns? it feels like the hash rates are so small that it wouldn't make a difference even with thousands of users online.
Thanks, Mamad
Assuming that they all leave their browser open for a continuous period at the same time; you could optimistically look at 1Mhash/s with 200 users (unless my math is off). And that is with an average of 5Khash/s per user, if you have a high portion of IE users or users with old browsers that will drop.
|
|
|
|
SgtSpike
Legendary
Offline
Activity: 1400
Merit: 1005
|
|
May 25, 2011, 06:15:09 PM |
|
First things first, I really love the idea. But I want to know if anyone is having any success with this. I just started running this on one site that has about 200 online users. What can I expect in terms of returns? it feels like the hash rates are so small that it wouldn't make a difference even with thousands of users online.
Thanks, Mamad
It won't. 1000 users at 1khps is 1mhps. It would take 270,000 online users, 24/7, to generate the same mhps that a 5850 does, stock.
|
|
|
|
mamad
Newbie
Offline
Activity: 17
Merit: 0
|
|
May 25, 2011, 07:24:18 PM |
|
Not to shoot down the idea, but after running it for about an hour on a busy site (250+ users online with average 5 minutes on site) I received 0 shares. So taking it down.
Still like the idea of having it run in browser, maybe ActivX would be able to use more CPU/ even GPU?
|
|
|
|
gigitrix
|
|
May 25, 2011, 07:47:52 PM |
|
Feature request: Can't you guys set this "subpool" to have a difficulty of less than 1 (0x00000000FFFF0000000000000000000000000000000000000000000000000000)?
If you set it at, say (0x0000FFFF00000000000000000000000000000000000000000000000000000000) then at least people would have a more even distribution. Hopefully this wouldn't overload the servers too much (tweak the values of course)
|
|
|
|
BitcoinsWallet
Member
Offline
Activity: 70
Merit: 10
|
|
May 25, 2011, 10:11:49 PM |
|
1bitc0inplz, disabled for now. Some users experiencing excessive slowness of their computers. If you implement one option for use some percentage (I wish use 50% of the CPU for example) and run it in more low priority side.... will be better And think that some users open a lot of tabs from the same website, be careful with that... Great work in anyway!
|
|
|
|
Turix
|
|
May 25, 2011, 10:50:25 PM |
|
1bitc0inplz, disabled for now. Some users experiencing excessive slowness of their computers. If you implement one option for use some percentage (I wish use 50% of the CPU for example) and run it in more low priority side.... will be better And think that some users open a lot of tabs from the same website, be careful with that... Great work in anyway! I'm currently using forceUIThreads and a long delay time to mitigate this somewhat (a delay of 100ms seems to be ~ 10% of 1 core on my E5200). But the hash rates are proportionally bad, my average hash rate is like 750hash/s right now.
|
|
|
|
cablepair
|
|
May 26, 2011, 12:28:51 AM |
|
Current average hash rate: 5188.021721523151
Shares this round: 2
so, how many mega or kilo hashes exactly is 5188.021721523151
what is it measured in?
|
|
|
|
lowentropy
|
|
May 26, 2011, 12:29:27 AM |
|
Feature request: Can't you guys set this "subpool" to have a difficulty of less than 1 (0x00000000FFFF0000000000000000000000000000000000000000000000000000)?
If you set it at, say (0x0000FFFF00000000000000000000000000000000000000000000000000000000) then at least people would have a more even distribution. Hopefully this wouldn't overload the servers too much (tweak the values of course)
I really like this idea! What does everyone else think?
|
Mine @ <http://pool.bitp.it> Chat with us @ irc://irc.freenode.net/#bitp.it Learn more about our pool @ <http://forum.bitcoin.org/index.php?topic=12181.0>
|
|
|
lowentropy
|
|
May 26, 2011, 12:31:23 AM |
|
so, how many mega or kilo hashes exactly is 5188.021721523151
what is it measured in?
This number is confusing and we'll be replacing it soon (read: hopefully before this weekend... we all have day jobs ) with better stats. Currently it's indicating the average speed of a given visitor to your site. So it looks like a good portion of your visitors are using fast browsers (safari, chrome, ff4, etc.). It's measured in just hashes per second.
|
Mine @ <http://pool.bitp.it> Chat with us @ irc://irc.freenode.net/#bitp.it Learn more about our pool @ <http://forum.bitcoin.org/index.php?topic=12181.0>
|
|
|
PIO
Newbie
Offline
Activity: 27
Merit: 0
|
|
May 27, 2011, 11:56:29 AM |
|
Just a quick update... We changed the default script that is generate for you to have the forceUIThread parameter set to true. This will force the script to be very, very, nice. You can still overwrite that setting by explicitly setting it to false, doing so will increase your hash rate at the risk of higher CPU usage for your website visitors. As promised, here is a complete list of parameters that can be passed into the engine: - clientId
- delay
- forceUIThread
- autoStart
- workerTimeout
clientId - this is your unique identifier. Do not change, unless of course you want your shares going to someone else delay - this is the delay, in milliseconds that the UI thread waits between every 100 hashes. This setting does not control web workers. Lowering this value too much could result in browsers displaying the "we've detecting that this script has locked up" warnings. The default is 30 ms. forceUIThread - some browsers don't support web workers (IE), in those cases this setting is meaningless. For browser that support web workers, this forces them *not* to use them, and instead calculate hashes in the UI thread at a much slower, and less CPU intensive, rate. autoStart - don't use this yet. Right now the jsMiner auto starts. You could set it to false. However, since the jsMiner is running in an iframe you would have no way of starting it... we will work on getting this better exposed. workerTimeout - this effects all browsers. This is the length of time (in seconds) that a get work request is good for. The default is 30 seconds. Increasing this value decreases network request but increases the likelihood of stale shares. And example of what you're modified script tag would look like with some of these parameters is: <script type="text/javascript"> bitpit({clientId: "5", forceUIThread: true, delay: 40, workerTimeout: 45}); </script>
Why dont you show the code on the website when we login?
|
|
|
|
PIO
Newbie
Offline
Activity: 27
Merit: 0
|
|
May 27, 2011, 11:59:18 AM |
|
Requesting mining stats in control panel Thank you Yeah, I dont even know if it's working or not.
|
|
|
|
1bitc0inplz (OP)
Member
Offline
Activity: 112
Merit: 10
|
|
May 27, 2011, 11:38:51 PM |
|
Why dont you show the code on the website when we login? That is a good suggestion. I'll add that to our list. Requesting mining stats in control panel Thank you Yeah, I dont even know if it's working or not. We are going to roll out even more stats this evening. But, do rest assured, if you used the JS we provided you it's working. But, once we roll out these new stats you'll be able to see for yourself.
|
|
|
|
Sergey (imcex.com)
Newbie
Offline
Activity: 20
Merit: 0
|
|
May 28, 2011, 06:34:58 AM |
|
We are going to roll out even more stats this evening.
Good job, guys! Thank you! Now, proceeding to level 2 Feature requests: 1. Some techsupport email or jabber to discuss personal issues in case of need 2. Stats columns: - khashes per Last Day - khashes this round - script requests per Hour - effectiveness = khashes this round / script requests this round (this will show how effective an average visitor for this site is) 3. Sorting by domain name/selected column. I guess client javascript sort will be a good solution. 4. Totals row - totals for each column where applicable, duplicate the row before and after the table. Thank you in advance.
|
|
|
|
1bitc0inplz (OP)
Member
Offline
Activity: 112
Merit: 10
|
|
May 28, 2011, 03:52:00 PM |
|
Good job, guys! Thank you! Now, proceeding to level 2 Feature requests: 1. Some techsupport email or jabber to discuss personal issues in case of need We already have support@bitp.it, plus you can reach us here, and even on Twitter @bitp_it We do need to make that more obvious, perhaps a contact us section of the site... 2. Stats columns: - khashes per Last Day - khashes this round - script requests per Hour - effectiveness = khashes this round / script requests this round (this will show how effective an average visitor for this site is)
3. Sorting by domain name/selected column. I guess client javascript sort will be a good solution.
4. Totals row - totals for each column where applicable, duplicate the row before and after the table.
Thank you in advance.
Those are all great suggestions. We are no where near done yet. We have some pretty big stuff in the pipeline.
|
|
|
|
Sergey (imcex.com)
Newbie
Offline
Activity: 20
Merit: 0
|
|
May 28, 2011, 06:03:12 PM |
|
We have some pretty big stuff in the pipeline.
It would be great, if you could publish some kind of roadmap. I understand that having little resources you got to make decisions between optimizing backend stuff and improving the frontend, but the priorities are not very obvious to other users (beta-testers). What I suggested in the last post - is the smallest bit of information that will help me decide whether it is worth using bitp.it for my own sites and where the bottlenecks are. Currently, I can only guess how effective this whole thing is So, please, roadmap of the upcoming features would be useful to keep patience
|
|
|
|
1bitc0inplz (OP)
Member
Offline
Activity: 112
Merit: 10
|
|
May 28, 2011, 07:02:03 PM |
|
So, please, roadmap of the upcoming features would be useful to keep patience You are right, time is a luxury, and we (like a lot of us on this forum) are doing this in our free time. We do not want to get into the habit of over-promising, and we also don't like the idea of over-thinking. It is our believe that good software is evolved, not designed. And while we do have a long term plan, the points between A and Z are somewhat undetermined. In fact, it is the feedback from people like yourself and everyone else that has replied to this thread that will help us pick the route between A and Z. But, I completely understand the need for a roadmap, so that people can make decisions. Let me put some thought into it, and put some things down on paper.... I will see what I can put up for a short-term road map.
|
|
|
|
1bitc0inplz (OP)
Member
Offline
Activity: 112
Merit: 10
|
|
May 29, 2011, 02:16:31 AM Last edit: May 29, 2011, 02:32:45 AM by 1bitc0inplz |
|
Today we finished up our first round of payouts. Everybody who earned 0.01 BTC or more has received their payment, with one notable exception. There was one account that earned a payout, but didn't have a wallet ID set... we are going to email you shortly and see if we can get your payment sent to you.
Additionally, everyone should now be able to see their pending payout in their console.
We have really been blown away by the level of response that bitp.it has generated. Never would we have imaged that the concept of mining bitcoins from your website would have been as popular as this has turned out to be.
However, there is still a lot of room for us to improve. We realize that even our top performers have received payouts that are lackluster. It would appear that too many people are still using slower browsers, computers, and just not staying on people's websites as long as we had hoped. We do believe that WebCL opens up new possibilities to web mining, but that technology is much too young. Hopefully, with time, it will gain browser acceptance, but until then it's just a 3rd party plugin for one browser that doesn't even work on 64-bit operating systems.
We are *not* discontinuing support of our Javascript miner. Actually, quite the opposite. We have a lot to do to grow this project into something bigger. Over the couple of weeks we plan on rolling out support for: - graph of past 10 days' hash rates - UI improvements - show total across domains - show average hash rate per domain
Additionally, we have something REALLY big planned for one week from today. We do not really want to elaborate what it is, but it will be a game changer for bitp.it.
Thank you all again for your support. We really appreciate you helping us fine tune this product.
|
|
|
|
AngelusWebDesign
|
|
May 29, 2011, 05:46:38 AM |
|
Maybe the websites that host this mining code could AT LEAST encourage its visitors to upgrade to a decent browser. Maybe a small graphic at the bottom of the page, linking to Chrome and/or Firefox? Or maybe throw in Safari and Opera as well. Such buttons would only appear if the user is browsing with some version of IE. It's true that Chrome is far, far faster than other browsers, notably IE and Firefox. I'm talking about general browsing performance, as well as Javascript performance. It's not just miners that want fast Javascript; doesn't everyone? About the only "downsides" with Chrome are for web developers -- but how many of THEM are out there, really, as a percentage of users?
|
|
|
|
AngelusWebDesign
|
|
May 30, 2011, 11:39:40 PM |
|
Here's what one of my users said (I had it in "forceUIThread: false" mode at the time)
I have a new computer and when I bring cathinfo up, it is very choppy and slow. Keystrokes lag and hesitate when typing. This does not happen on any other website. Is there something I can do about this?
|
|
|
|
|