Scrypt Jane Research - Post Ideas Here
-
[quote name=“ghostlander” post=“14920” timestamp=“1371232995”]
GPU mining will come to demise just like CPU mining has come already. You can’t beat inevitable. We need a solution of painless transition to ASIC mining. A controlled by community solution to avoid all mess we see happening now with Bitcoin. I’m ready to help any honest developer who wants to design and market a scrypt ASIC without having the community left behind.
[/quote]ASIC makes crypto available to all. The price points are great and the distribution to all is possible. It’s the zombies who don’t want to produce anything only jump currencies like rabbits on crack that I worry about. ASICs in the hands of Gold 2.0 lunatics will cause bitcoin to suicide down to $20. Guaranteed.
[quote]
[quote]Coblee believes that there is only room for one coin under each hashing algo. Any other coin will see its difficulty rollercoaster which seems true for us right now as we swing between 41% swings up and down.[/quote]That’s what he says. Litecoin has suffered lately due to Feathercoin and a few other altcoins. Some of them are to disappear soon, but Feathercoin is here to stay and they don’t really want the competition. We’re less than 2 months old and look at our achievements!
[/quote]He wants the “Me too” market. Well that’s what he got. GPU tweakers.
[quote]
We can surely implement a different kind of scrypt, but I’m inclined to think it isn’t going to bring us any real advantage at this moment. Our losses may be higher than gains.Bitcoin won’t change its hash function to scrypt. Not now, not ever. Even if their developers are crazy enough to approve this decision, people who have invested money into ASICs are to stay on the old block chain. ASICs are contributing already to over 60% of their hash rate including 24% by ASICminer alone.
[/quote]I really believe multiple layers is going to be the way to go. Without multiple layers we will have constant market overheat and crash patterns because people are thinking short term. If we want to support commerce, we have to begin there. There is no gradual step from ADHD to smooth cruising.
-
I’m going to stare at a folder of folders and put folders in the folders. Time to move.
-
[quote name=“zerodrama” post=“14930” timestamp=“1371233844”]
I’m going to stare at a folder of folders and put folders in the folders. Time to move.
[/quote]What does this mean, and besides that you like to see yourself post, what is the point of you posting this?
-
[quote name=“randomdef” post=“14945” timestamp=“1371236031”]
[quote author=zerodrama link=topic=1839.msg14930#msg14930 date=1371233844]
I’m going to stare at a folder of folders and put folders in the folders. Time to move.
[/quote]What does this mean, and besides that you like to see yourself post, what is the point of you posting this?
[/quote]It means I’m going to rearrange the folders and code in LeatherCoin so people can quickly find the code they seek to do tests etc.
-
Well since we’re debating scrypt/scrypt-jane/sha256, let’s talk openly here.
Why would people bringing huge investment to the coin be bad? FPGA/ASIC miners are not cheap to design or manufacturer, and there’s still the power investment.
If people could suddenly invest in Feathercoin by buying specialized hardware, wouldn’t this ultimately be a good thing?
Bitcoin maintains it’s price because people have a vested interest in seeing it do so… so it does.
It seems to me we’re having a problem of people not being invested enough in it’s value to continue to mine it, wouldn’t making a change that would allow people to take their already considerable investment in Bitcoin mining, and allowing them to get on board be a GOOD thing?
What’s the argument for keeping it specialized, and not commercially available again? I seem to have missed it.
-
[quote name=“Kevlar” post=“15018” timestamp=“1371242752”]
What’s the argument for keeping it specialized, and not commercially available again? I seem to have missed it.
[/quote]SHA256 is like sitting on an infinite pile of wet noodles to achieve security. And every advance means you have attackers with 100s of times the speed of everyone else.
SHA256 generates constant paranoia and prevents evolution to commerce.Do we actually have 19 MILLION diff worth of transactions per second? This is like the story of the guy who tried to outrun a locomotive. OVERKILL.
Scrypt is 1000x as strong so you only need a little bit to achieve the same degree of security. Scrypt also gets technological advances with a more reasonable 5 to 15x boost. You don’t have to keep reinvesting in hardware to do continue doing business.
Scrypt-jane if done right blows away attackers. And these zombies are coming.
I’m sorry, but we cannot base decisions on pure numbers. These numbers don’t just go up and down. They accelerate and stagnate. They destroy investment that has yet to be paid off.
I constantly see people trying to make numerical arguments free of philosophical content and I’m sorry but that’s the road to failure.
Again, I would be for ASIC IF and ONLY IF the ASIC crowd were not a bunch of Gold 2.0 lunatics.
-
Zerodrama: A well reasoned reply. Thank you.
Let’s keep the debate going. :)
-
[quote name=“zerodrama” post=“15050” timestamp=“1371244533”]
Again, I would be for ASIC IF and ONLY IF the ASIC crowd were not a bunch of Gold 2.0 lunatics.
[/quote][img]http://cdn.memegenerator.net/instances/300x300/38756083.jpg[/img]
[img]http://cdn.memegenerator.net/instances/300x300/36869211.jpg[/img]
sorry… levity…
-
[quote name=“zerodrama” post=“14929” timestamp=“1371233778”]
I really believe multiple layers is going to be the way to go.
[/quote] -
[quote name=“jeremiel” post=“14912” timestamp=“1371231876”]
I’ll play devil’s advocate.Let’s go over the scenario of literally changing the algorithm. What’s the process everyone would have to follow(miners and users) once it’s changed? New client? Restructure of pools? New miner deployed? A start date for change over?
[/quote]Did we miss this post? ^
I am curious on the steps this would need to be streamlined and seamless for such a large change.
-
[quote name=“Smoothie” post=“15106” timestamp=“1371251569”]
[quote author=jeremiel link=topic=1839.msg14912#msg14912 date=1371231876]
I’ll play devil’s advocate.Let’s go over the scenario of literally changing the algorithm. What’s the process everyone would have to follow(miners and users) once it’s changed? New client? Restructure of pools? New miner deployed? A start date for change over?
[/quote]Did we miss this post? ^
I am curious on the steps this would need to be streamlined and seamless for such a large change.
[/quote]No one answered it. The current conversation is about the justification for moving to a new algorithm.
-
[quote name=“jeremiel” post=“15145” timestamp=“1371266899”]
[quote author=Smoothie link=topic=1839.msg15106#msg15106 date=1371251569]
[quote author=jeremiel link=topic=1839.msg14912#msg14912 date=1371231876]
I’ll play devil’s advocate.Let’s go over the scenario of literally changing the algorithm. What’s the process everyone would have to follow(miners and users) once it’s changed? New client? Restructure of pools? New miner deployed? A start date for change over?
[/quote]Did we miss this post? ^
I am curious on the steps this would need to be streamlined and seamless for such a large change.
[/quote]No one answered it. The current conversation is about the justification for moving to a new algorithm.
[/quote]I personally think this should be a part of the discussion, if it isn’t feasible then no point in looking to make modifications. I’m not against change, just feel we need a plan.
-
[quote name=“Smoothie” post=“15163” timestamp=“1371275770”]
[quote author=jeremiel link=topic=1839.msg15145#msg15145 date=1371266899]
[quote author=Smoothie link=topic=1839.msg15106#msg15106 date=1371251569]
[quote author=jeremiel link=topic=1839.msg14912#msg14912 date=1371231876]
I’ll play devil’s advocate.Let’s go over the scenario of literally changing the algorithm. What’s the process everyone would have to follow(miners and users) once it’s changed? New client? Restructure of pools? New miner deployed? A start date for change over?
[/quote]Did we miss this post? ^
I am curious on the steps this would need to be streamlined and seamless for such a large change.
[/quote]No one answered it. The current conversation is about the justification for moving to a new algorithm.
[/quote]I personally think this should be a part of the discussion, if it isn’t feasible then no point in looking to make modifications. I’m not against change, just feel we need a plan.
[/quote]I agree hence why I asked the questions. Yes we are talking about a software change but what would be required by the rest of the community to get onboard with a new “possible” format. Sort of a fire drill. I know talking about could make people perceive it’s actually happening besides it being concept implementation plan.
Also, it’s friday… people take breaks on friday from the internets sometimes…
-
Remember that we are just debating this issue right now and that nothing is set in stone.
I will speak to Coinotron about what he might need to do on his end. Going with Scrypt-Jane I would commission Coinotron to do what he needs to do and then test it on my testnet. Whatever solution for the pools we come up with will be shared with the community well ahead of time. Perhaps the client can switch between hashing and the pools do not need to do anything, I’m not sure about this. I’ve spoken to Coinotron briefly about this already and he did not put forward any objections.
The biggest challenge is getting the miners to move over. After the change all their Scrypt shares will become rejected until they restart with the new algo. If we can get cgminer to support it then miners could close, rename Scrypt to Scrypt-Jane in their config then fire up their miners again.
This is all doable but it is going to cause a lot of work for everyone involved. I do not know of another coin that has changed its hashing algo like this so we have nothing to compare against.
-
[quote name=“Kevlar” post=“15018” timestamp=“1371242752”]Bitcoin maintains it’s price because people have a vested interest in seeing it do so… so it does.
[/quote]
That vested interest takes many forms. My partners folks invested 1000’s of $ in to BTC. Some folk want to see a change to FIAT markets. Others like the concepts of what these currencies bring to markets.Guess what I am saying is, is that ASIC chips as a form of investment is not the main thing holding the value up… far from it.
Personally I’m here at FeatherCoin because of ASIC mining. The ASIC resistant claim sold this concept to me. It told me it was fairer and available to more folk wishing to get involved at the grassroots level.
I don’t for a single minute think I have a strong argument on the real pro’s and Con’s of ASIC mining… but ASIC set ups will not be available to all in any sense imho and thus making FeatherCoin a less accessable coin if it went down the ASIC route.
-
Bushstar,
Thanks for addressing that point.
I will have to reread this thread to get a better grasp of the details of the new hash algorithm etc to understand for myself what would need to happen in the scenario that we go forward with this.
Perhaps we can list the pros and cons and discuss it that way also.
-
I am confused to as this is technically feasible. Wouldnt changing the algorithm break all the previous calculations? IE everyone would have to start fresh?
-
No, the previous calculations are not recorded. Verifying a block is different from creating one. The clients and miners including transactions only need to verify.
-
Thank you Bushtar for responding. Everyone needs to know this is all concept but even implementation could be painful for everyone. In the end I see this coming down to mining software. Though there might be a bit of time where the network hashrate is 0… that could be scary.
-
[quote name=“Smoothie” post=“15237” timestamp=“1371293542”]
Perhaps we can list the pros and cons and discuss it that way also.
[/quote]Sounds like a good idea, I am more interested in the cons of Scrypt-Jane to see what challenges we might face.
[quote author=jeremiel link=topic=1839.msg15363#msg15363 date=1371315222]
Thank you Bushtar for responding. Everyone needs to know this is all concept but even implementation could be painful for everyone. In the end I see this coming down to mining software. Though there might be a bit of time where the network hashrate is 0… that could be scary.
[/quote]Advanced Checkpointing could help in the move to make sure no one starts replacing the chain after the move. We would also need some people to hash against Scrypt-Jane before the switch over, all these shares will be rejected until the network hits the new hashing algo block.
A lot of testing needs to go into this. Later on if we are still interested in this and there are no show stoppers then I will run a Feathercoin Scrypt-Jane test pool with a fast difficulty adjust so together we can test this solution and see how it works in practice. We can trial run the change over from one algo to another on the test network.
We do not want to leave anything in this process to chance.
