Well, I have to say something about this pseudo "Project" called Bitcoin Utopia...

It seems they now hook into ASIC based miner hardware that is very specific in application and can't be used for anything but bitcoin mining.

First of all, I always thought of and think of BOINC projects as being based in humanitarian and research projects of all types. Thinking this way causes me to believe the following though I know there is no formal BOINC statement to this effect.

1. Hooking into existing bitcoin hardware is in NO WAY related to humanitarian/research projects. It's simply a way to concentrate $$$$ generation in one node... Bitcoin Utopia.

2. Bitcoin Utopia itself is NOT a research project of any kind.

3. The Horribly Imbalanced points awarded for this project will PERMANENTLY upset the real world BOINC COMBINED statistics and may thus have the longer term effect of removing some measure of motivation to participate.

4. The ASIC hardware is BITCOIN specific which abrogates the general concept of "more or less equal for all multipurpose PC hardware". No other project can use the bitcoin ASIC hardware for anything but bitcoins.

While the intent to generate project funding is good, the points awarded amount to the same thing as BUYING BOINC points. Look at these outputs!!

http://www.bitcoinutopia.net/bitcoinutopia/top_users.php

I think Bitcoin Utopia should be excluded from BOINC COMBINED totals.

The same Blah Blah Blah Blah when the GPU's came out ... You Guy'z should Crunch More & Think Less, it won't hurt your Brain's so much if you do that ...

If you're crunching for humanitarian reasons, why do you care about credits?

Bitcoin Utopian raises REAL MONEY for science used to help humanity. It fills gap in Boinc projects. Think of it like taxes. There's nothing humanitarian about taxes, but the money raised is (sometimes) used for humanitarian aims. (Note to self: Maybe the government should give out credits and badges based on taxes paid ).

BU has cpu, gpu and ASIC projects. Anyone can crunch it.

Finally, other projects are completely free to RAISE the amount of credits awarded. Wouldn't that entice even more humanitarian crunchers to participate?

Isn't learning to apply new technology (ASICS), to supersede a formally antiquated method of supporting research (taxes/donations), a form of research in and of itself?


See my first reply.


The same argument was made when GPUs were first utilized on BOINC projects. IIRC it was either PrimGrid or Collatz that first implemented GPUs. And note that GPUs only lend themselves to highly paralleled forms of data processing. Many BOINC projects, by their very nature, are exempt from using GPU tech.

And what about projects like Radioactive@Home and the family of Quake-Catcher Network projects? All these projects require specialized hardware that is not ubiquitous within the realm of PC hardware.



Well, isn't that the point? After all, how many BOINC projects will ever see their research actually applied to solve the issue(s) that prompted their formation in the first place? The answer is... no one really knows. But, BOINC volunteers crunch them none the less.

Anyhoot, to all who live in the States, Happy 4th of July!!!

Oh and a belated Happy Canada Day to all my neighbors to the North.

Completely false. Before you go on a rant you should do your homework. These devices hash the SHA-256 algorithm, which was in use long before Bitcoin was ever even thought of. It is a method of encryption, plain and simple. On top of that, mining Bitcoin is a byproduct of their actual purpose when used in this manner, which is to verify the legitimacy of Bitcoin transactions. After every Bitcoin has been mined, this is something that will still have to be done.


How is raising money to donate to scientific endeavors not humanitarian? Seems to me you're one of those people who'd rather complain than spend a measly $15 to get in on the action and help fund science which you claim to care about so much.

Since i already covered your first 2 complaints, i guess it would break your heart to know those of us who use ASICs on Bitcoin Utopia are horribly underpaid credit-wise. The regular ASIC workunits should actually pay 460,800 credits apiece and the large workunits should pay 3,686,400 credits apiece. This is based on the real-world amount of actual work done by an ASIC compared to a GPU. We gladly accept this severe reduction in credits to keep people like you from crying about it in public forums, and yet here you are...


OMGNOWAI! It's an outcry! People being granted credits based on the amount of work done! It's the end of the world! We can't have fairness around here...what were they thinking! :O Buying BOINC points? Let's see, so i mined some fraction of a bitcoin which was turned over to BU. But then BU gives the money away too so...who are we paying again? And what about you? You use your computer for BOINC points. I guess you're buying credits too by your logic. Afterall, you're paying for the hardware and electricity to make it happen, saving the various projects from having to pay for it. In return for the money you pay out, you get BOINC credits as well.


Every member of BOINC in the world has had exactly the same amount of time to get the proper hardware and figure out how to use it. Don't come here with your whinging because you feel you have no idea how to get involved. Life is about learning.

If you're trying to be "competitive" in any aspect with BOINC, you'll have to sink a lot of money into hardware. If you need the competitive edge, compare yourself to team members or friends, if you go at it on the open membership list you will set yourself up for disappointment. I would just appreciate that fact that you're doing what you can to help researchers, and encourage others to do so as well. Even though you may not be dumping as much money and hardware into the BOINC projects as others are, your contribution of resources IS important, regardless of how powerful. After all, that's kind of why grid computing got started in the first place... To leverage the power of as many machines as possible, in order to reduce the need for dedicated and specialized computing equipment.

It will be interesting to see how crypto-currencies play out in the long run, will the technology keep up with the mining needs? If not, the profitability will disappear and so will a lot of miners who are interested in the profits, not supporting an open currency.

John, I'm not weighing in on my opinion for or against the OP's opinions. I will point out that you have a flaw in your statement here. BU does in fact get paid because they take 12% (was 15%) of all Bitcoins mined by volunteers. Slice it how you want, but the truth is that they are taking their cut.

Maybe we should set in Radioactive@Home 100k points per Wu?

I'm just kidding - don't shut me, please

Yes that is true. That money is used to pay their electric bills and server upkeep, not to put in their pockets as profit. BU reduced their fee from 15% to 12% when more people started getting more workunits done, as you said. This was in spite of the fact their electrical and hardware costs increased many times. This is the same use as any donation given to any project or even to a stats site like this one or Free-DC. Other projects like Milkyway@Home use donated money to pay certain staff members.

If you don't donate the money directly, chances are they have a grant of some sort. The money from government grants is raised by taxpayers. In one form or another, we are all handing over money to keep BOINC projects afloat, even if you never crunch a single workunit. At least when you crunch for BU or make a direct donation to any project, you have a say where your money is going, instead of leaving it all in the hands of some grant committee.

ASIC is one kind of specific devices, so it can not be used in any other science projects. The stats from BU are just showing their strong mining power. I want to see the REAL science computing power, not mining power. Bitcoin Utopia should be excluded.

Then you should exclude radioactive and all the quake projects. They require project unique hardware too.

I completely agree with Tex1954 in this project should not be allowed to give so much credit per work unit. Being it is a fund raising project it should be categorized differently. Even donate@home didnt give out that many points. Seti@home USA made almost 5 billion credits/day, Sicituradastra almost 3 billion and Team Musketeers with 1.4 billion. Its not a matter of if this project should be allowed or the moral of a fund raiser project this is about the stupid amount of ppd people can get and its only ruining other projects due to lack of support.

The simple math of it, rather the the feeling on what points should be:

I have been told that a Boinc developer has indicated to the BU project that a boinc credit has been valued at 4.3e11 FLOPS (1 credit = 4.3e11 FLOPs, BU forums for reference). The following is based on that number.

For BU, the larger jobs labeled with 20+ GH/s ASICs require at least 10240 shares to be completed and pays 600,000 credits.

Two examples:
1) It took a AMD HD 7970 14 secs to crunch 1 share of bitcoin as measured by running 7970 on BU and looking at the time it took to do the share, not the overall workunit time (which has setup and reporting time. A 7970 is 3.8 TFLOPs.

10240 shares x 14 secs = 143360 seconds for a 7970 to do a 20gh+ wu for BU.
3.8 TFlops x 143360 secs = 5.44768e17 Flops used for the wu by a 7970
5.44768e17 Flops/4.3e11 (one credit) = 1,266,902 credits per 20gh/s+ wu

This math indicates we are already at half value for the wus.

2) 7970 vs Rockminer R-Box - Modified example from user Bryan over at BU forums
Some sites derive the equivalent FLOPS of an ASIC device by comparing the hash rates of 2 devices.
The AMD HD7970 GPU will mine BTC at a 700Mh/s rate. Rockminer R-Boxes will hash at a 35Gh/s (35,000Mh/s) rate which provides 50X more computational power than a HD7970.

A Rockminer R-Box can produce a completed 20+ GH/s Asic wus at BU in 19 to 25 minutes. Lets use the avg 22 minutes (1320 secs). When I average my last 190 results from a single R-Box I get 22.11 minutes and my R-Box is overclocked.

7970 3.8 TFLOPS x 50 = 190 TFLOPS equivalency for R-Box (1.9e14)
1.9e14 FLOPS x 1320 = 2.508e17 FLOPS for a R-Box to do a 20GH/s wu.
2.508e17 FLOPS / 4.3e11 (one credit) = 583,255 credits per 20gh/s+ wu

Not too far off of what the project is paying (2.8% difference). It is interesting that in the 2 examples the FLOPs needed are such a wide range, probably due to the small sample side of the 7970 wu and the equivalency really only being a way to estimate the FLOPS for the R-Box. The amount of work done and the credits that should be awarded is somewhere between those two based on the 4.3e11 FLOPS = 1 credit metric that was given to the project by a developer.

If I bought 50 7970s and did the work, everyone would be ok with that number of credits? What difference does it make that I use a device that is cheaper to do the same thing? I am pretty sure we have never used a cost of equipment to credits granted system. Yes, this rate of credits upsets the apple cart. But that is evolution. If we want to change the credit to FLOPS ratio that is used, all projects would have to adjust to the same ratio. wouldn't they?

Basing the credits on the time to do work v.s. amount of work done (measureable in FLOPS) doesn't seem right to me, there would be no incentive to help the projects go faster by using newer equipment. It would just be silly if a Pentium and a 7970 got the same credits, if the 7970 did 10 times the work in the same time period. Same logic applies here.

If any of the math above is wrong, please let me know and I will fix it.

Regards, Em

If you follow the official credit rule, you will get ZERO. I suggest BU and QCN projects should move to the "Specific/Special-Purpose Hardware" area. Any projects can create their own credit rule in that area and get what they want.

Just show you the correct math using the official credit rule.

SHA256:
1 hash = 6350 integer operations
1 hash/s = 6350 IOPS (Integer Operations Per Second, not FLOPS)

Bitcoin Utopia & ASIC:
32GH/s R-BOX = 32e9*6350 IOPS = 203,200 GIOPS
450GH/s Rocket BOX =450e9*6350 IOPS = 2,857,500 GIOPS
Do you see the GFLOPS ? I think no, ASIC miners can not do float-point operations.
203,200 GIOPS but 0 GFLOPS (2e14:0)
How many credits should be given? http://boinc.berkeley.edu/wiki/computation_credit
GigaFLOPs = RAC/200 or RAC = GigaFLOPs*200
By the credit rule, 0 GFLOPS should get 0 credit/day.
Zero is absoultely right, because ASICs can not be used in any science projects.
　
Donate@Home & GPU:
600MH/s AMD HD7970 = 600e6*6350 IOPS= 3,810 GIOPS
3810 GIOPS implies 3810 GFLOPS (1:1)
By the credit rule, 3810 GFLOPS should get 3810*200 = 762,000 credits per day.
Is 762,000 reasonable?
Yes, GPUs are very important and often used as accelerators for many science computations.
We should give so high credits for GPU crunching.
　
　

Thanks for your reply. First let me says that I do agree that asics don't do floating point operations. However, it was the Boinc developer that was originally working on the equivalency for BU and the basis for formula used in my post, but unfortunately there were a few issues with his original attempt (off by a multiple of 1,000 and he received bad values from a forum thread). These things happen.

Your idea that work units that are done by asics should get 0 is innovative. But it implies that a cpu that crunches on BU gets credit, a gpu that crunches on BU gets credit, but if the same work is done by an asic device it gets 0 credit. It also implies that all projects doing integers operations and not floating point need to get 0 regardless of the device type. Your statement that ASIC devices can not be used on science projects is not true. An Application Specific Integrated Circuit (ASIC) has limited scope, but the technology as a whole does not. They can and are developed for many uses. The thought that the mining devices cannot help science projects is also not quite true, any project in need of funds can employ the same means to raise funds and further their scientific goals. That is what Donate did and what the BU project does for three different scientific endeavors (one of which is the Boinc project MilkyWay). Field-programmable gate-arrays (FPGAs) which are used for many of the same purposes as ASIC are also much faster than GPU/CPU, I suppose they shouldn't be allowed either. Here is a nice primer https://www.princeton.edu/~achaney/tmve/wiki100k/docs/Application-specific_integrated_circuit.html

Am I understanding your last part, you think that Donate was ok because a GPU did the work. Is that correct? What you say about GPUs is also true of ASICs, they are very important and often used as accelerators. If it is correct that Donate was OK but BU isn't because inaddtion to CPU and GPU it allowed for of the advancement of ASIC, Sorry I don't follow that reasoning at all. Also if you are claiming x IOPS implies x FLOPS, I fail to see the problem with the examples I provided.

if GIOPS implies GLOPS
35Gh/s Rockminer r-box * 6350 IO = 222,250 GIOPS
Credits per day 222,250 GIOPS * 200 = 44,450,000
Which is more than rockminers earn in a day now


If you want to use the lower end of an r-box
32GH/s R-BOX = 32e9*6350 IOPS = 203,200 GIOPS
Credits per day 203,200 * 200 = 40,640,000 which is still more than a Rockminer r-box earns today.

If Bitcoin Utopia IS only a device to HELP BOINC projects, it shouldn't award points at all in the general scheme of things...

Because in that case, strictly speaking, it isn't a BOINC research project...

Shouldn't we give folks POINTS for donating money to a project? Why not? Because it didn't happen to burn a bunch of electricity on a computer setup?

Same thing... it's a good idea and helps, nothing wrong with it, but then why can't I get points for donating cash? What's the difference?

But I have donated to several projects and never got a single damn point... although I got a couple goodies like a T-Shirt or lanyard... LOL!

Is that fair? Sheeshh... And then comes the AMOUNT of points awarded thing... I could care less about that, I agree the ASIC's are faster than GPU's, but that isn't the point at all; it isn't a real BOINC project!

Bitcoin Utopia is in fact a "fund raising device" and shouldn't give points at all... not unless I get points for donating real CASH... It doesn't research directly or contribute directly to any problem "solving" at all... However, some will argue the whole BOINC platform idea never excluded use by anybody for anything they wanted or could do with it.

Soo, I stand by what I originally said... Bitcoin Utopia should be EXCLUDED from BOINC Combined stats...or failing that, given a separate category page or something somewhere... Perhaps it's time to separate BOINC combined into categories...

Perhaps it's time to block posting for people with ZERO credits.

Another cowardly haters...

Looks like our 0 points poster Tex1954 is quite active on BOINC.
LINK to his STATS.

http://boinc.berkeley.edu/trac/wiki/CreditGeneralized

^I support the above.^

Combined credit will still be available. This will preserve the BOINC competitiveness as we "knew" it.
I'm sorry to see the hard feelings in this thread. Passionate bunch I guess.

Would love to hear what Willy has to say about grouping credits by type; this may require some retooling.

New credits system from DA.
http://boinc.berkeley.edu/trac/wiki/CreditGeneralized

The most idiotic sentence:


If you stupidly wasting electricity by using a CPU for mining, then you do "science" and then you can receive normal credits.

If for the same you use the most efficient ASIC, then you do not "science" and then you can not receive normal credits.

Now BOINC will not be a place for science, but a place for total hypocrisy.

After many years of supporting it is time to move away from BOINC.
I can use my miners for my money and I do not have to subsidize any boinc project - no money for MW.
I can use my GPUs for Folding or something else and I do not have to support any boinc project.
I can use my CPUs e.g. for prime searching outside boinc...
I can not support BOINC - system that blocks the technical development and promotes wasting energy.

BOINC is a platform for volunteer computing. That's the original definition. Where does it say science? Who decides which projects are fit to run on BOINC ? answer no one, volunteers run BOINC an open source program available to all. If you don't agree with a project , don't run it , simple. It's been well proven that BU fits with the BOINC credit regime. The fact that it is not a ' science ' project is irrelevant. There have been and are many non ' science ' projects on BOINC. There are many projects in BOINC that, to some are very important and to others are a waste of electricity. People should stop trying to impose ' their ' views on others and accept that we are all volunteers and can run any project we like, science or not.

http://boinc.berkley.edu the home of BOINC together with its associated wikipedia explains what BOINC is. Please note that nowhere does it states that it is or it was designed purely for Science Research. Although, apparently it is ideally suited for scientific endeavors there have been many projects that had nothing to do with scientific endeavours. Therefore those who maintain that BU is not a Science Research project are correct but it has nothing to do with why they should not be included in any stats.

As a Volunteer BOINC follower I have a choice as to which project I crunch. The money I invest in purchasing up to date technology is my choice also. What I can not understand is why is it necessary to change yet again our credit system or the way it is reported. It was my choice not to crunch Seti@Home and others under the Credit New system for it prevented me to catch up to the top users (Yes, I am highly competitive) who got there by receiving higher scores than I can get under the New Credit system. Credit parity does not work and discourages people using up to date technology that can perform a task faster than a reference machine. Non BOINC projects can give high credit in appreciation for the work you performed using up to date technology and for an equivalent time for similar work I can only get a few points. I joined Folding@Home less than 24 hours ago and I have already received over 17,430 credit for using just a couple of cores on one machine. My suggestion would be to increase the base reference machine's point system to grant MORE credit rather than less and set a high ceiling. Let the volunteers decide like in supply and demand as to which project they wish to follow instead of being controlled centrally like in a Communist system. Ultimately, it will be the volunteers who will decide the fate of BOINC for without Volunteers it will not survive. Unfortunately, we don't have much say in this for it is the developers who will decide our way of doing things without asking what we as volunteers think about it.

Why is it necessary to change the way we categorize our results. There are many sites that promulgate results and some are even combine Non BOINC project results in their stats. They have a combined table, a table for Individual results, Team Results, Highest individual scores and scores for team results, scores for best machines, graphs and more. All you have to do is look at the results of your choosing and what is offered by the Stats Sites. If what you are looking for is not there than look at another stat site. I do not like the idea of dictating to Volunteer Stat sites what to include and how it should be presented by any central authority. If DA thinks this is necessary to break up BOINC into 8 or 10 different compartments, than I suggest that DA and the BOINC Developers create an OFFICIAL BOINC Stat site that is funded and run by the central authority. It is my choice as to which stats site I will support and donate as to its upkeep. Forcing it on us and dictating by a central authority is not for me. I believe in free choice. Leave the Volunteer Stat sites alone, allow them to
promulgate results according to users wishes rather than a centrally controlled one.

Well, I am in the middle of house remodeling (slow process) and also go out on the road for months at a time. As of now, I can't leave my systems running while on the road so credits come and go...

However, I did want to point out that Robert7NBI made a typical (fallacious) ad hominem statement; this Robert7NBI person attacks "ME" instead of the issue I present.

In any case, I stand by my arguments. Truly, I don't see the "NEED" to keep a BOINC Combined in any sense. We could exclude it entirely! It is only these stats websites that track combined points anyways. "WE" created the "combined" points system, "WE" can decide we don't like it anymore or not.

I'm done with this now, said what I wanted to say. Like others, BOINC will always be important to me, especially cancer and related research projects. Nobody gets a trophy for being top on BOINC Combined that I am aware of other than a certain prestige. I think all attempts to make it one thing or another will be fruitless now.

I'll continue to do what I have done... run what I want... and continue to ignore BOINC Combined stats leaving that to the folks that see BOINC as some sort of contest.

Happy BOINCing to all BOINCers!

8-)