Topic: These new EFFICIENT x11 algos everyone is talking about ?? BULLSHIT or real? - page 10. (Read 16302 times)

Every algorithms in X11 is able to do ASIC since day one because it is requirement from NIST to become SHA3 candidate.

No, it won't, it's just a matter of the software.
For the moment the only method to secure against ASICs is a combination of algos for which there are no ASICs.

A real rat race. But funny. I like it.
An opportunity to call out people for their true motives.

I honestly wonder: why hasn't anyone created yet an algo that randomly changes parts of itself every x blocks? This would make ASICs impossible, and only FPGA possible to use, thus limiting the issue a lot...

Quark's algo is the best I've seen. Sadly, I don't know if its worth a major investment because I'm not sure if the community can overcome the FUD from months ago or if the community can figure out how to market itself to people who will buy it.

x11 is mainly a marketing ploy.

Would people be interested in an IPO for my x50 coin? Much secure

Ah, great you're back again. Thought we'd lost you for a moment... so now you're back and want to waste more time. Can you take a look at these and clear up these points you made. I am genuinely interested in understanding your answers to the questions i asked you before.... here's where we were at.

Hmmm interesting...let's take this step by step... you don't give clear answers so i will try to clarify what you are saying.


1. No x11 is not more efficient than qrk. Yes we appreciate the efficiency is due to the mining software and the gpu.

2. Is x11 more secure than quark ? you say yes with no explanation.... expand on this. Secure against what attacks? and how is it more secure?? (answer unclear at this time)

3. In your opinion with some justification the miner may not be all that crippled. Perhaps you are correct we don't know. Really 50 % gpu only and that is fully optimised?? Maybe there is a serious bottleneck somewhere and further optimisation is not possible... but i guess nobody knows for sure yet. Even you can't know for sure can you?

4. wow NO... as simple as that?  really, explain how you can guarantee this fact? we do need some reason to back up opinion, perhaps you have a good reason but really you didn't express it all that well.

5. Hmm does having more algos essentially make it more asic resistant. Does the fact these are sequential and therefore predictable not make it easier for asics than qrk since that is random?  I have no idea on this.. perhaps others can expand on this one. Either way are both x11 and qrk less asic resistant than memory hard algos?

6. No it is not more asic resistant that scrypt N. Since it will be most costly in terms of hardware to implement scrypt N asics. So it seems scryptN is more asic resistant due to costs? or you consider the costs of the extra hardware not as much of an issue as creating an asic to tackle the 11 algos? 

7. x123 ....... should we fork to this immediately or build up one at a time ? Smiley

Thanks, I really appreciate an honest opinion.
I always had a heart for the disabled.

I'll try to decipher that .... however, thanks for your input so far you have been quite helpful.

I won't waste my time with you anymore.
It's a lot like listening to a try of a conversation between two parking lot hookers.

CU later

thank you, yes i am enlightening myself also.... we are only trying to find the truth and yet such hostility ensues ....

There is no agenda here for my part aside from the fact i don't want to risk hardforking 3 coins i am mining now to x11 if it is not even worth doing so and there are better options.

Hmmm interesting...let's take this step by step... you don't give clear answers so i will try to clarify what you are saying.

1. No x11 is not more efficient than qrk. Yes we appreciate the efficiency is due to the mining software and the gpu.

2. Is x11 more secure than quark ? you say yes with no explanation.... expand on this. Secure against what attacks? and how is it more secure?? (answer unclear at this time)

3. In your opinion with some justification the miner may not be all that crippled. Perhaps you are correct we don't know. Really 50 % gpu only and that is fully optimised?? Maybe there is a serious bottleneck somewhere and further optimisation is not possible... but i guess nobody knows for sure yet.

4. wow NO... as simple as that?  really, explain how you can guarantee this fact? we do need some reason to back up opinion, perhaps you have a good reason but really you didn't express it all that well.

5. Hmm does having more algos essentially make it more asic resistant. Does the fact these are sequential and therefore predictable not make it easier for asics than qrk since that is random?  I have no idea on this.. perhaps others can expand on this one. Either way are both x11 and qrk less asic resistant than memory hard algos?

6. No it is not more asic resistant that scrypt N. Since it will be most costly in terms of hardware to implement scrypt N asics. So it seems scryptN is more asic resistant due to costs? or you consider the costs of the extra hardware not as much of an issue as creating an asic to tackle the 11 algos?  

7. x123 ....... should we fork to this immediately or build up one at a time ?

I will rename the "CPU Only" topic to CPU Friendly or something similar. The current, publicly available X11 GPU miner does NOT have a large advantage over the X11 CPU miner. That is why it is why X11 coins are on the list.

The current GPU miner may leave room for optimization.  

Here beggar, just to dismantle your house of crap I'll give you some details about the facts you don't understand.

1. is it more efficient than qrk? - There is nothing like efficiency to the algo itself. Software could be efficient. The algo is the algo.

2. is it more secure than qrk?

Yes

3. is it more efficient than scrypt? or is the miner just crippled and can't use the full potential of the card?

It does use 100% of the intruction set of available GPUs at a very good level of optimization.

4. are there more efficient miners already out for x11 that are more optimised?

No.

5. is it more asic resistant that qrk?

Yes, it uses 11 algos, Quark only uses 6. It could change to anything else tomorrow.
Fear mongering of "FORKING" is just a matter of deployment of the right software.

6. is it more asic resistant that scyptN , scrypt jane

Of course. The only advantage these methods have is an increased amount of memory necessary to run the algo.
That means higher power consumption and extreme utilization of the hardware, for nothing actually.
It doesn't improve the security of the challenge, it just takes more power to solve.
So it's, in my opinion, a step in the wrong direction, as is the assumption to make a coin more secure by the size of the network.

You`re trying to sell camel shit on a camel market here.

7. is it x12, x13, x99 going to be better? will we need to fork all of the coins over and over again?

No, I think X123 will be the final solution.

50% might be the right amount, but I didn't find x11 to be the most efficient out of all algorithms
groestl so far is the most efficient on my computer (both the myriad kind and the groestlcoin kind)
I don't know how much more optimization those miners can get, though

there is nothing special about x11 at all

@cryptohunter

You cant beat the trolls and you cant educate them. You are doing a great job with this thread to enlighten the community, but dont drive yourself crazy. 

They may well use x11 but before they decide to they just need answers to a few of these questions.

1. is it more efficient than qrk?
2. is it more secure than qrk?
3. is it more efficient than scrypt? or is the miner just crippled and can't use the full potential of the card?
4. are there more efficient miners already out for x11 that are more optimised?
5. is it more asic resistant that qrk?
6. is it more asic resistant that scyptN , scrypt jane
7. is it x12, x13, x99 going to be better? will we need to fork all of the coins over and over again?


just yes or no with some kind of reasoning they can follow.... it can't be that hard can it to answer a few questions to the best of your knowledge?

If we had a direct answer to some of those backed up with some real data that could be a great starting point ...

You could make it run much faster on a farm with a smart paralleling solution.
That would be a thread dispatcher for the algos among entities delivering a timely result according to the total cycle time.
Get a quote from the guys who do the weather forecast software.
Or from Disney, or lets call a name, PIXAR.

Or Google.
Whats up with the Google Coin. If it's for real I'd expect a real overcast for anything Altcoin.
They probably come up with their own algo.

Or maybe they'll just use X11, bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha bwahaha.

Please troll, give your noob mouth a rest. We are talking here, don't want your continued screaming and shouting about asic supporters?  i see none in here.

This has nothing to do with asic support. Go back and read the entire thread. Then get your parents to read over it and assist you with your comprehension skills.  

flipme and slapper..... are you two morons a couple?

Thanks for pointing this out. The ASIC MOB is pulling all registers now! Watch out.
Stupid, uneducated posters, street smart crooks, name calling, tabloid phrases.

You got anything substantial to add to your own thread OP?

OK so it seems that some algos will bottleneck certain parts of the gpu before others so since other parts are left redundant it draws less energy. I did actually ask that exact thing earlier in the thread but nobody took up on it. Okay so let's be fair... so algos on x11 could be running as fast as they ever will on our gpus and will burn less energy than scrypt? is that a fair assumption?  

However is 50% sounding correct? that does sound like some miner tweaks and mods could be very useful..the bottlenecks seem rather large here?

Also since none of these algos in this mix or qrks mix are memory hard, it seems those 2 should be equally as efficient since the algos will be creating these same bottlenecks and stoping the memory getting over taxed? or very very close too close to seperate really. So x11 is not ground breaking with regard the efficiency aspect since we can see sha256 and all the other algos in QRK and x11 are possibly just as easy on our electricity usage.


So the super new efficiency story seems to be a good marketing  at best or perhaps just wishful thinking.... but perhaps not as open to super tweaking of the miner as we thought could be taking place.

So we rule out going x11 based on efficiency alone...


Let's do it like this.... a kind of battle of the algos....


so for efficiency running on gpu with current miners - - any algo that is not memory hard?  qrk, x11, single algos ...sha 256 ... all the others...


so for asic resistance - - scryptN, vs scrypt jane - high n, vs QRK, vs X11.   Let's do that one now.... who wins the anti asic battle and why? technical difficulty to create and cost to create...lets take in all the factors here....??

the very first SHA-256 miner was only maybe 5% less efficient than the best one (poclbm was giving me good hash rates before a lot of these newer kernels optimized it by a few more percent)
people just found 1% improvements here and there a few times and after a while there were no optimizations left
some optimizations reduced the total instruction count by one out of 1300 (less than 0.1% improvement)

SHA-256 takes like 20% less energy to mine off the top of my head because it doesn't use the GPU ram as much
so I'm sure there are algorithms that take even less GPU power because they rape one part of the chip (bottleneck) and don't touch the rest of it much