Maybe that might help a bit:
Algos extracted from SPH-SGMiner sources
https://bitcointalksearch.org/topic/ann-sph-sgminer-multi-coin-multi-algorithm-gpu-miner-added-marucoin-475795- FC FuguCoin (1 time 256) R9 290 145MH/s
1 fugue256
- Ink INKcoin (2 times 512) R9 290 45MH/s
1 shavite
2 shavite
- MYR MyriadCoin (2 (512+256)) R9 290 10MH/s
1 groestl
2 sha256
- GRS GroestlCoin (2 times 512) R9 290 5.5MH/s
1 groestl
2 groestl
- Q2C Qubit (5 times 512) R9 290 5MH/s
1 luffa
2 cubehash
3 shavite
4 simd
5 echo
- SIC Sifcoin( 6 times 512) R9 290 4.5MH/s
1 blake
2 bmw
3 groestl
4 jh
5 keccak
6 skein
- DRK DarkCoin X11 (11 times 512) R9 290 2.5MH/s
1 blake
2 bmw
3 groestl
4 skein
5 jh
6 keccak
7 luffa
8 cubehash
9 shavite
10 simd
11 echo
- QRK Quark (9 times 512) R9 290 2.1MH/s
1 blake
2 bmw
3 groestl OR skein
4 groestl
5 jh
6 blake OR bmw
7 keccak
8 skein
9 keccak OR jh
- ANI AnimeCoin (9 times 512) R9 290 1.8MH/s
1 sph_bmw
2 sph_blake
3 groestl OR skein
4 groestl
5 jh
6 blake OR bmw
7 keccak
8 skein
9 keccak OR jh
----------------------------------------------------
Sorted for Hash/s
(Workload per Hash[1/H] : Speed[MH/s] -> worklaod*speed -> rel. Performance)
- FC FuguCoin ( 256 : 145 -> 37120 M/s)
- Ink INKcoin ( 1024 : 45 -> 46080 M/s)
- MYR MyriadCoin ( 768 : 10 -> 7680 M/s)
- GRS GroestlCoin ( 1024 : 5.5 -> 5632 M/s)
- Q2C Qubit ( 2560 : 5 -> 12800 M/s)
- SIC Sifcoin ( 3072 : 4.5 -> 13824 M/s)
- DRK DarkCoin X11 ( 5632 : 2.5 -> 14080 M/s)
- QRK Quark ( 4608 : 2.1 -> 9677 M/s)
- ANI AnimeCoin ( 4608 : 1.8 -> 8294 M/s)
Sorted for relative Performance:- Ink INKcoin ( 1024 * 45 -> 46080 M/s)
- FC FuguCoin ( 256 * 145 -> 37120 M/s)
- DRK DarkCoin X11 ( 5632 * 2.5 -> 14080 M/s)- SIC Sifcoin ( 3072 * 4.5 -> 13824 M/s)
- Q2C Qubit ( 2560 * 5 -> 12800 M/s)
- QRK Quark ( 4608 * 2.1 -> 9677 M/s)
- ANI AnimeCoin ( 4608 * 1.8 -> 8294 M/s)
- MYR MyriadCoin ( 768 * 10 -> 7680 M/s)
- GRS GroestlCoin ( 1024 * 5.5 -> 5632 M/s)
-----
Acording to this raw comperation X11's performance is on the better half compared to the other algos in tose miner.
Of course this doesn't really takes the differences between each Hash algo into account and the size it takes on the GPU
So it is a raw comperation but an interesting one since all those Algos are build out of the
very same Hash Moduls in the miner and used the very same way.
Now it can be that all the Hash Moduls are not very efficient and a comperation with other miners that have the same has moduls but implemented in a different way would be very interesting!!
Specially interesting is that the sph miner has the keccak hash but no MaxCoin implementation !?
The CudaMiner has a good and fast keccak so it had been very interesting to compare this! I wonder if it's maybe because the
single keccack modul isn't that good implemented in sph-miner?
I don't have much experiences with SPH-miner nore all this Algos with exception of X11. So maybe others can sheed more light on this.
---
Personel I don't really bother the efficiency or non efficiency of X11.
For me the point of X11 is simple (I discovered it a week ago):
I finnaly can get home and don't have to endure 50 degree celsius in my house and a noise that you can't enjoy anything anymore!!
And it uses less power. It's like a dream for one that simple wants to mine a bit. The standard joe if you want so.
And the summer is comming fast.
As long as the majority has the same ..maybe inefficient miner... we all have a big advantage from it!
If one figures out that you can optimize this to get it hot and noisy as scrypt..if thats even possible..I think it's not or not to that degree..which would mean that the standard GPU isn't optimal for those algos..then I guess some like me simple will search for the next cooler "inefficient" algo.
Security and such is secondary since ..well you know... how long do you hold a coin as miner?