Author

Topic: CCminer(SP-MOD) Modded NVIDIA Maxwell / Pascal kernels. - page 1051. (Read 2347601 times)

legendary
Activity: 3164
Merit: 1003
[...] the amount of vram isn't a bottleneck in any algo, rather memory bandwidth [...]

With Scrypt-Jane at high n-factors (and possibly with Scrypt-N too, that one I have no experience with), greater amounts of VRAM make a huge difference. Sidenote: only for Linux, since Windows sux and can't ever fully utilize the VRAM.
I suppose these are somewhat the exception, since most other algos don't really care for memory quantity at all.
Yes the latest versions of windows does... but my vista does 3 x what windows 8.1 does on hashrate with Scrypt-Jane  16.
legendary
Activity: 3164
Merit: 1003
The 980 ti
http://www.gigabyte.com/products/product-page.aspx?pid=5208#ov

Would there be a difference if the card where 6gb and not 4gb ram on hashrate ?
thx
to answer the question, just look at how much memory you use on your card, if you are close to the 4gb limit, might help if not well it won't...
(the amount of vram isn't a bottleneck in any algo, rather memory bandwidth)
actually memory bandwidth is higher, so that's part might improved a bit things up  

anyhow, probably better to wait, there are some rumors which say that the price will go down a bit...
That's the problem ...I can't wait much longer.  Smiley  But thx djm
legendary
Activity: 3164
Merit: 1003


The source got released, but he has done a private neoscrypt kernal that does 750khs/s on the 980 up from 630.

My free opensource sp-modded version is only around 7% faster than 630. around 680khash

now doing 820kh/s on the 980
               496kh/s on the 780
               196kh/s on the 750ti (my 750ti was just doing 165kh/s before...)
Good improvement. Smiley
How do I get the updated miner djm?
you pray  Grin
I contributed....do you want more beers.  Grin
legendary
Activity: 3164
Merit: 1003
The 980 ti
http://www.gigabyte.com/products/product-page.aspx?pid=5208#ov

Would there be a difference if the card where 6gb and not 4gb ram on hashrate ?
thx

That's a non-Ti.

This is a Ti which have 6GB memory: http://www.gigabyte.com/products/product-page.aspx?pid=5472#ov



Yes thx  I thought I was looking at a 980 ti with 4GB and that's why I got confused. Embarrassed
legendary
Activity: 1154
Merit: 1001
Thought monero also uses a lot of memory.

Relatively speaking, I think it does, but with Monero, I haven't seen any difference between similar 2GB and 4GB cards. So while it uses a lot of VRAM, it cares more about memory bandwidth, less about absolute memory quantity (my experience is only with 2GB up).

Obviously, the developers will know much better, but my basic understanding is that with Monero, you just need enough VRAM to fit a certain/large scratchpad. Beyond that, more memory does not help in any meaningful way. Faster memory does help.
There might be coding optimizations/tradeoffs to be made, much like larger -L values do for Scrypt-Jane & Scrypt-N (not storing the full scratchpad and computing on the fly as needed for missing values).

Hope the description makes some sense. If not, may the bashing begin  Grin
legendary
Activity: 1764
Merit: 1024


The source got released, but he has done a private neoscrypt kernal that does 750khs/s on the 980 up from 630.

My free opensource sp-modded version is only around 7% faster than 630. around 680khash

now doing 820kh/s on the 980
               496kh/s on the 780
               196kh/s on the 750ti (my 750ti was just doing 165kh/s before...)

What is the hash/watt stock clocks? Really should be included with any advertisement of hashrate at this point.

[...] the amount of vram isn't a bottleneck in any algo, rather memory bandwidth [...]

With Scrypt-Jane at high n-factors (and possibly with Scrypt-N too, that one I have no experience with), greater amounts of VRAM make a huge difference. Sidenote: only for Linux, since Windows sux and can't ever fully utilize the VRAM.
I suppose these are somewhat the exception, since most other algos don't really care for memory quantity at all.

Thought monero also uses a lot of memory.
legendary
Activity: 1154
Merit: 1001
[...] the amount of vram isn't a bottleneck in any algo, rather memory bandwidth [...]

With Scrypt-Jane at high n-factors (and possibly with Scrypt-N too, that one I have no experience with), greater amounts of VRAM make a huge difference. Sidenote: only for Linux, since Windows sux and can't ever fully utilize the VRAM.
I suppose these are somewhat the exception, since most other algos don't really care for memory quantity at all.
legendary
Activity: 1400
Merit: 1050
The 980 ti
http://www.gigabyte.com/products/product-page.aspx?pid=5208#ov

Would there be a difference if the card where 6gb and not 4gb ram on hashrate ?
thx
to answer the question, just look at how much memory you use on your card, if you are close to the 4gb limit, might help if not well it won't...
(the amount of vram isn't a bottleneck in any algo, rather memory bandwidth)
actually memory bandwidth is higher, so that's part might improved a bit things up  

anyhow, probably better to wait, there are some rumors which say that the price will go down a bit...
legendary
Activity: 1400
Merit: 1050


The source got released, but he has done a private neoscrypt kernal that does 750khs/s on the 980 up from 630.

My free opensource sp-modded version is only around 7% faster than 630. around 680khash

now doing 820kh/s on the 980
               496kh/s on the 780
               196kh/s on the 750ti (my 750ti was just doing 165kh/s before...)
Good improvement. Smiley
How do I get the updated miner djm?
you pray  Grin
legendary
Activity: 2002
Merit: 1051
ICO? Not even once.
The 980 ti
http://www.gigabyte.com/products/product-page.aspx?pid=5208#ov

Would there be a difference if the card where 6gb and not 4gb ram on hashrate ?
thx

That's a non-Ti.

This is a Ti which have 6GB memory: http://www.gigabyte.com/products/product-page.aspx?pid=5472#ov


legendary
Activity: 3164
Merit: 1003


The source got released, but he has done a private neoscrypt kernal that does 750khs/s on the 980 up from 630.

My free opensource sp-modded version is only around 7% faster than 630. around 680khash

now doing 820kh/s on the 980
               496kh/s on the 780
               196kh/s on the 750ti (my 750ti was just doing 165kh/s before...)
Good improvement. Smiley
How do I get the updated miner djm?
legendary
Activity: 3164
Merit: 1003
The 980 ti
http://www.gigabyte.com/products/product-page.aspx?pid=5208#ov

Would there be a difference if the card where 6gb and not 4gb ram on hashrate ?
thx
legendary
Activity: 1400
Merit: 1050


The source got released, but he has done a private neoscrypt kernal that does 750khs/s on the 980 up from 630.

My free opensource sp-modded version is only around 7% faster than 630. around 680khash

now doing 820kh/s on the 980
               496kh/s on the 780
               196kh/s on the 750ti (my 750ti was just doing 165kh/s before...)
legendary
Activity: 1484
Merit: 1082
ccminer/cpuminer developer
For me cpuminer was the first project working on both vstudio and gcc (linux + mingw), even if i had to create the vstudio project. I learnt a lot with that. I only made small projects before in Visual C.

Then i moved to ccminer to learn cuda... I work on these two project since one year at almost 100% of my time. I had also to setup dependencies and a MinGW64 build env.
sr. member
Activity: 462
Merit: 250
I just would like make my peer aware of an auction for six 750TI 2GBs at this forum. So far bid is 1.51 BTC.  Shocked

Link: https://bitcointalksearch.org/topic/computer-hardware-selling-6-gpus-nvidia-750tis-2gb-asuspalit-mobo-4-free-1098900

legendary
Activity: 1764
Merit: 1024
I still say you can build a company off of a percentage based miner at 2%. While my figures were off as I based BTC prices at $500 a pop instead of $200 due to lack of sleep, it's still $3k per month JUST for Quark JUST on Nicehash. There is plenty of money to be made across all the algos, especially when you dump AMD into the mix.

Yeah, but as Wolf0 said it would be just a matter of time before the donation part would get cut out.

But... now that yaamp source is public someone could create a custom multipool where you can only connect with a custom miner.
The miner wouldn't have code for donation, simply the pool would take a fixed donation.
For protection, a new version of the (free) miner could be pushed out let's say every few weeks with hardcoded changes while also changing the pool so that only the newest version of the miner could connect.
The periodical changes should discourage people trying to hack the miner unless I missed something.

That's where DRM comes in. It only gets cut out because developers for kernels really don't spend any time on the actual program itself. They're all modified versions of the same miner all the way back to the beginning. Activation, encryption, verification... all things that could be done, but would require development time. Of course something a company can tackle. Continued support and updated versions with rotations on all the above keep people in the loop.

At a low enough donation level, people don't take time or want to cut it out either. Just the same as people buy legal programs, people are willing to pay for software as long as it's reasonable. All of this put together would keep such things to a minimum. The examples of reverse engineered miners involve quite large fees.

And yup custom mining pool, updates. There was a groestl AMD custom miner that did that. Long story short, there are plenty of ways to deal with this and keep people mostly legit. It's not about getting everyone, it's about discouraging people from doing it and making it not worth their time. Added features, benefits, constant updates, feeling that the developers care also keep people honest because they like that sort of thing. That may be too much to tackle for one person, but not for a company.
hero member
Activity: 672
Merit: 500
No, I definitely didn't read your post as a recruitment. It was just a bit opaque.
member
Activity: 97
Merit: 10
So basically what you're writing is:

thank you for your for your work guys. Have you considered helping even more?

No not at all. I did say casually help out and I don't expect them to do anymore then what they already have done, there is no obligation to any of these guys so I don't know why it was read this way sometimes words can be misguided I suppose. I know none of these guys actually get paid for this as its not like a day job for them or a career move. They are probably already IT professionals working as programmers or doing a degree at university and maybe partaking in some projects as a hobby/interest in cryptocoins and maybe learning a few things on the way too. I would say this would also look very impressive on a resume too Smiley

I was simply asking as I was curious if they are interested in any development in coins or just in mining software aspect. These guys are talented and I think would be a great asset to a dev team with all these improvement they have made, that's all I was trying to say.
Maybe they might some day make their own alt coin with some new blockchain features never seen before that could become a game changer, interesting times we live in no doubt.
In case your wondering I have no affiliation with any coin development team and zero experience in coding, not trying to recruit anyone in case you accuse me of this  Kiss
hero member
Activity: 672
Merit: 500
So basically what you're writing is:

thank you for your for your work guys. Have you considered helping even more?
member
Activity: 97
Merit: 10
Every algorithm has been optimized some are 200% faster than Christian's original ccminer code (blake).

I have over 666 commit's to github the last 10 months. I have spendt 1000's of hours... Smiley

same, maybe more hours.. and no... you didnt took blake from christian ccminer... i wrote it... and djm reused that for lyra2

git log blake32.cu

didnt know when i did it that it was also in cudaminer :p

Well its heroes like you 2 that make mining worthwhile, truly amazing work being done lately with all this mining software, without the software being updated for new hardware and drivers constantly emerging, alt coins would be dead without the talents of people like you guys. Have any of you considered joining any of the devs for your preferred coins to casually help out?
I know Maxcoin team for example has had a few big coin reward bounty's for projects that nobody has claimed yet.  They really need an android wallet its been on the cards for some time but there is little interest it seems.
Jump to: