 my average is 640kh/s using this setx GPUMAXALLOCPERCENT 100 setx GPUUSESYNC_OBJECTS 1 del *.bin --gpu-platform 0 -d 0 -w 256 -v 1 -I 18 -g 1 -l 1 --shaders 1792 --lookup-gap 2 --scan-time 1 --queue 0 --thread-concurrency 21712 overclock is 1150 core 1660 vram how you getting 700+ without a ton of errors? What brand/model of card are you using? I'm using MSI 7950. Of the brands I tested, MSI was the only one which could be pushed this far. Sapphires died around 680. Here's my bamt config: "gpu-fan":"100", "intensity" : "20", "vectors" : "1", "worksize" : "256", "kernel" : "scrypt", "thread-concurrency" : "24000", "expiry" : "120", "gpu-engine" : "1145", "gpu-memclock" : "1500", "gpu-powertune" : "20", "log" : "5", "no-pool-disable" : true, "queue" : "1", "scan-time" : "60", "scrypt" : true, "shares" : "0", "kernel-path" : "/usr/local/bin" Also using a modified kernel. Without the kernel, those settings will push an MSI to 672-685. Can't say what it'll do to other cards. MSI uses the best quality components, you can push them hard. These cards have been running since September. There was 1 30 minute rest period as they were moved from one location to the next. I had 2 MSI cards eventually die, and MSI paid me $285 each per card. Not bad considering I paid $200 to buy the card new to begin with. Again, MSI, best warranty. |
|
|
|
|
My MSI 270's and 270X's run 493 stable, with some 499/500+. It's possible.
|
|
|
|
The same could be said about you and your 1 card. This is big boy land. Play hard. Do your best with your 1 card. A 7950 should be mining at 715kh.
I have never heard of a 7950 get 715khash. Most I have heard of is 650, and I can get up to 642 on my 7950, and maybe the occasional little uprising. Can you show a reference getting 715?  That was taken a few days ago. Same system is still running with 0 issues. Now imagine racks of them. ASRock 970 EX4, Sempron 145, 2GB ram Seasonic 850 MSI 7950's BAMT 1.3, cgminer 3.7.3, Modified scrypt kernel Engine: 1145 Memory: 1500 Power/Intensity: 20 Each card pulls around 240w. |
|
|
|
|
Sounds like starcoin lol.
The zombie coin
|
|
|
|
|
Voted best pool op of 2013!
|
|
|
|
Are there any pools working
Are you drunk sir? Best post in the whole thread. |
|
|
|
if you've got such a large system that 18hrs isn't enough, i pray to god its not. that way i can at least get some mining in on my SINGLE 7950 before you slaughter my hopes and dreams of making enough coins to buy another card using this coin.
I started out with a 40kh card a few months back. I now have a small, but growing, farm which uses close to 100amps of AC power 24/7. Work hard, and it will pay you back. you bigger rigged out people need to calm your tits. try putting yourself in a broke persons shoes, like mine!
I can honestly say I know how you feel, but again, I reinvested my earnings, got more equipment, kept doing it.. now the system pays for itself. bunch of cry babies here.. get your kleenex, call your mommies and get off the forums till you're done being babies.
The same could be said about you and your 1 card. This is big boy land. Play hard. Do your best with your 1 card. A 7950 should be mining at 715kh. gpucoin changed some things based on what you all wanted or brought up valid points for, and you're going to say "its going to be a rocky start". i'd much rather mine for someone thats willing to work with the community than someone that says "its my way and you dont matter".
Having been involved in launches, there's a reason I'm telling him his last minute changes could make things really bad.
I've been trying to get one on Amazon for 2 months. They get like 5 in a month for the normal $329.00,
what you are seeing is all the 3rd party merchants on Amazon, and they are never under $450
I have many windows open, each one to a GPU, and refresh is set to 3 seconds. They do pop up on prime, and they disappear literally in seconds. I saw 500+ MSI cards go in 10 minutes. I was one of the folks notified that they had oversold Sapphire Toxic cards in a matter of seconds. They got their monthly truck shipment in last week, and the ASUS cards sold far slower than anything else. A few hours ago they had normally priced "Prime" cards. The MSI's and Sapphires vanished like steam. Might I suggest you do what I do? Keep the windows open, check constantly, set email alerts. I even wrote custom scripts to email/text-me/wake me up when they come in stock, so I could hit the "buy it now" on my large order. And when you buy by the crate, they sometimes even come wrapped in cellophane! |
|
|
|
I agree, it is evident he is doing what he sees as best, and I am still excited about the coin despite not agreeing with everything. Whether I'm right or wrong, I just want to see the coin and business model become valuable and longlasting. 24Kilo, take a look at this: http://www.reddit.com/r/BAMT/comments/1ylhhy/release_vertcoin_bamt/I don't mind going to PM and helping you.  |
|
|
|
Needless to say, I fully agree with vesperwillow (really great analysis and time-line of facts) and ny2cafuse (another bunch of dev team facts).
Mark, this is how you managed to loose the trust of the community. Moreover, you are acting as "king", saying I won't respond here - if you want to talk with me then come to my kingdom (Tagbond). I will be available for conversation only there. I'm to good to respond here.
Reading you last post Mark really makes me laugh. The situation is getting ridiculous and of course the price of TAG is falling.
I'm not a fan of any conspiracy theories, but things written by vesper and fuse are a way closer to the truth, in my opinion, than the things you say, Mark.
I think you owe a huge explanation to the people how supported you for so long. For now, it just seems you don''t give a rat's ass for them.
K.
Yeah, it's amazing what dots become clear when you keep notes and start laying them out on a table. And when the price hits 1 satoshi, what margin will he have for profit? Less than he has now I'm sure. And if the whole community here stops, he loses 1/4-1/3 of his hashpower, so his overhead increases. Maybe my speculation is all off.. but hey, it's more information than we've received on this coin lately. In future I will be responding only via our forums at tagbond.com - if anyone wants to continue conversations I will respond there only.
Which currently contains 1 post by you in December announcing its' existence. http://forums.tagbond.com/discussion/12/tagcoin-general-discussion and is no doubt moderated. Sorry Mark but that is unacceptable (as if you really care). The peasants are revolting and instead of coming out to meet them, you choose to hide within the walls of your castle. I've seen the light and have walked away and I'm sure a lot of other people feel the same. Good luck with your broken coin. Lol.. nice catch man. |
|
|
|
|
Dabs, nice man! I love the b/w effect btw. Sorry about your PM! I've been backlogged and I don't look at the inbox often. I'll look toight/tomorrow.
|
|
|
|
Because the crypto world needs another LTC clone, right? We haven't watched enough of those come and go already? It's not that hard to set up your rigs to mine scrypt-n, and NO, it doesn't damage anything.
As for your previous comment: YES this should help stop a forkfest on the launch. Scrypt-n means we won't have multipools dropping 50ghs on it as soon as it's compiled. Would you rather have an instamined clusterfuck?
It wouldn't be a clone. Litecoin had no backing, it's value came because it is better for small transactions. Most clones are just dumps. The value in GPU Coin is set at the store level by the owners despite what anyone thinks. If it settles at 1 satoshi, they price their cards accordingly. It's that simple. The business model is the backbone to this, not the coin. Clusterfuck or not, it would've settled out. It always does. Good launches are so vital to a coins' popularity and value, which is why it's boggling they'll change things so late in the game rather than just switch over. And yeah it's not hard to setup systems for ScryptN. Honestly, it seems more lazy to ask for something which is resistant to a technology. Whynot invest your earnings into the new technology? I can think of only 2 scrypt coins which were based on a business model which have and continue to hold value. Guess what? It didn't matter who mined it or how, or how many orphans occured. As long as transactions processed, the business front end made it valuable. You can't expect to stand up a business and please both the miners and the other customers (folks who are simply going to buy the coin low at exchanges in order to get a good deal on the equipment from the store--folks who arent mining). If you want your business to not collapse on itself, you will please the website customer. This negates the miners' feelings. Every other coin to come along and try to please the miners and the front end in a business model, has failed. Business is a cold hard bitch. |
|
|
|
I'm an IPO myself, and know the coin would work fine with Scrypt. Like many other coins. Litecoin sound familiar?
I guess we won't know for another 2.5 years (amount of time litecoin has been out). In 8 months ScryptN ASIC will be out. So I'd say less than 2.5 years. The point of releasing the source code to pools is so they're ready to roll. Holding the passwords would prevent that. The other points about forking are also valid, but remember guys, you chose ScryptN. Very few farms out there with it right?  Exactly my point. Pools shouldn't BE "Ready to Roll" If someone wants early coins then solo for a few. Not like its hard. Could prob teach a monkey to solo if you had a fairly smart monkey and some patience. Also soloing helps to spread nodes so that the pool wallets dont fork the chain. Not like I expect to get any coins while soloing a hyped coin like this but if it is not done then watch this baby fork till the cows come home and then write the coin off. waltsmith The other side of the coin is a lot of coins have issues with launch because the distribution point becomes DDOS'd. Valid concerns on both sides I agree, but he's wagering either way. This is why a launch with standard Scrypt and then if necessary a move to a new algo, would've been better than changing half his original gameplan less than a day before launch. |
|
|
|
I'm an IPO myself, and know the coin would work fine with Scrypt. Like many other coins. Litecoin sound familiar?
Should work fine with just scrypt. This is a coin based on a business model difference, not a technological superiority.. in that sense the algo can be anything and still achieve the same effect. Look at auroracoin.. ^-- this guy gets it. I said the same thing to GPU Coin. |
|
|
|
|
I'm an IPO myself, and know the coin would work fine with Scrypt. Like many other coins. Litecoin sound familiar?
|
|
|
|
The point of releasing the source code to pools is so they're ready to roll. Holding the passwords would prevent that. The other points about forking are also valid, but remember guys, you chose ScryptN. Very few farms out there with it right? |
|
|
|
My intentions when GPU Coin was first devised was to provide a path to future mining, which I had assumed what the community wanted were ASICs, but was woefully wrong. After many corespondents on the subject I decided to drop the idea entirely.
When a business doesn't know what the community wants it needs to change in order to provide the people with the best support possible.
You took a few-hours poll on a thread on a bitcoin forum about what a GPU community wants in a future altcoin, and decided that was your majority input? Wow. You realize the most valuable alt coins out there aren't ScyptN, nor are they going ScryptN, for a reason..right? I took input, then did research on my own. Nothing is valuable until people put value on it. I personally think diamonds are expensive chunks of broken dreams, but that doesn't stop people from paying tens of thousands of dollars for them. You cite not wanting a bad launch, but changing things up in the 13th hour isn't good. Obviously I think it's a bad move, but it's your coin, and that's fine. We'll see how it goes. Hope it launches well. |
|
|
|
My intentions when GPU Coin was first devised was to provide a path to future mining, which I had assumed what the community wanted were ASICs, but was woefully wrong. After many corespondents on the subject I decided to drop the idea entirely.
When a business doesn't know what the community wants it needs to change in order to provide the people with the best support possible.
You took a few-hours poll on a thread on a bitcoin forum about what a GPU community wants in a future altcoin, and decided that was your majority input? Wow. You realize the most valuable alt coins out there aren't ScyptN, nor are they going ScryptN, for a reason..right? |
|
|
|
|
GPU Coin, as an investor, I also sent you my input on the debate.
|
|
|
|
Invested 0.5BTC for IPO.
I called for a N-factor.
For those of you don't just because you don't wanna reconfigure your rigs, you are gonna have to anyway when the 28nm Scrypt ASIC hit.
Why? Some of the most valuable coins aren't ScryptN, and never will be. And ASICs are coming for those new algos. Trust me on this. It's not impossible. I can only speak about 280x's since that is my expertise, and I have owned many of them.
NOTHING was more solid than the Asus direct CU top 280's, not MSI, not Sapphire..Maybe
previous Asus products were not as good, but the cooler on the CU is the best as well as
I guess it falls under YMMV. I ordered crates of all of them. Literally all brands and models. I kept what worked best under linux. I tracked settings, power use, etc, from underclocked/stock/overclocked. ASUS faired average among them. Sapphire and MSI always came out on top, with MSI being the most stable (less than 0.1% flux) and TOXIC was next in line with about 1% flux. And I'm not sure about "previous", I'm talking in the past 2 months. the chips being cherry picked for the top line. And are you kidding? Asus direct CU top 280's
are NEVER in stock ANYWHERE online for more than 2 seconds.
Really? Because amazon has a few hundred which they've had for the past 50 days. And newegg has them. Just checked. Unless folks dont buy from those places |
|
|
|
ASUS gpus are horrible, plenty of threads of horror about them. Please find other brands, even PowerColor.
MSI makes the best 270's (mine are 492-510kh), TOXIC makes the best 280's. Rock solid stability. Get a bunch of those and you're golden.
Amazon sold literally 2 trucks of MSI's out in a matter of less than an hour. ASUS are still plentiful..
You have NO idea of what you are talking about. I returned two Sapphire's that sucked. Asus 280x CU TOP is THE BEST hash and coolest/quietest 280x made, PERIOD. +1 I haven't heard any negative reviews about the Asus 280x, except the stnd DOA's that every manufacture has once in a while. While I like the toxic's the best, the Asus 280x is voltage unlocked which makes my life way easier. I have 6 280x DCT's and lol to vesper, PowerColor, really? If you haven't heard any negative reviews, you must have your head in the sand.. ASUS has been dogged like a stepchild on lots of mining forums, reddit, etc, because of a multitude of issues. It's not just a random string of failures. When tigerdirect, amazon and lots of other places have plenty of ASUS cards still in stock, while all the OTHER brands are sold out right now.. that should tell you something. That, by itself, should absolutely tell you something. And yeah, the powercolor I've drieid ran cooler, faster, and used less power, than the asus cards I tried. It's not my preferred, but it did a hell of a lot more than asus did. The 280 and 290's might have fewer issues, but they certainly do have their issues. They simply aren't robust. MSI is rock solid. Stout performance, takes it up the yang no matter how hard you push them or how hot they get (best components in the industry), and best warranty in the industry. Focusing on just one factor, sure ASUS are probably ok. But there are better options. Sapphire TOXIC is probably the sickest 280 card out there. Best hashrate whether undervolted or not, and superior components as well. |
|
|
|
|
Show Posts
