Pages:
Author

Topic: [ANN] DERO: DAG + Cryptonote + Bulletproofs + SSL + POW + Smart Contracts - page 49. (Read 123288 times)

full member
Activity: 152
Merit: 100
The problem of lost trust extends into any long term planning and capital investment. Who will buy an ASIC knowing the algo can be switched overnight? Who will buy coins if reward can be changed without any warning?

And the question about regulatory pressure on incorporated Dero is valid too. Looks like many questions and concerns have been raised and not yet addressed by the team.
newbie
Activity: 237
Merit: 0
Am I late, the main network is already up?
Tell me how many transactions per second it processes?
Hopefully more than" old " blockchains.

Yes, pools already works on new blockchain. Now it looks good, if you want to mine DERO you can do it on fairhash.org
newbie
Activity: 237
Merit: 0
v1 blocks cannot be mined (these are imported  blocks), rejecting

Can someone explain me?

Algo is still cryptonight, nothing change here
full member
Activity: 420
Merit: 115
Am I late, the main network is already up?
Tell me how many transactions per second it processes?
Hopefully more than" old " blockchains.
newbie
Activity: 12
Merit: 0
v1 blocks cannot be mined (these are imported  blocks), rejecting

Can someone explain me?
newbie
Activity: 237
Merit: 0
Guys, find with https://dero.fairhash.org/ super fast blocks on a new blockchain

Nice thank you, decent website btw...a lot of pools have crappy sites, so thank you :-)

we will be glad to see you and other miners here, as well, join our community in the telegram channel https://t.me/joinchat/B-GpfU1DYnzi5mhQ45I3rQ
full member
Activity: 420
Merit: 184
You could produce a secret hybrid ASIC/FPGA around $500 easily: You have to put every Hash functions use by CryptoNight like keccak, Jh, AES, Groestl, on an ASIC chip, and use a low fpga with some SRAM for the main loop and the scratchpad.

If Bitmain already produce the sophon sc1 for AI, why they couldn't produce a similar one for mining like above ?


You can certainly offload the hash functions to an ASIC, but if there is a need to move a lot of data into and out of memory rapidly then you either have to suffer with performance no better than a GPU (and that's only if your board design engineers are as good as those at Nvidia or AMD) or else put it on the ASIC die, which is exceptionally expensive (there's a reason why there is so little L3 cache relative to system DRAM). So the performance of this setup would likely be better than a pure FPGA solution, but hard-coding the hash functions into an ASIC is what makes an ASIC so easy to brick.

That said, I'm no longer interested in fighting this battle. The DERO team decided to stick with original CN because reasons and neither I, nor anyone else, is likely to change their minds. That doesn't mean I'm going to go buy an ASIC, however; the network will just have to get along without me, which I'm sure it will do just fine.

newbie
Activity: 49
Merit: 0
Guys, find with https://dero.fairhash.org/ super fast blocks on a new blockchain

Nice thank you, decent website btw...a lot of pools have crappy sites, so thank you :-)
newbie
Activity: 237
Merit: 0
Guys, find with https://dero.fairhash.org/ super fast blocks on a new blockchain
jr. member
Activity: 111
Merit: 1
You could produce a secret hybrid ASIC/FPGA around $500 easily: You have to put every Hash functions use by CryptoNight like keccak, Jh, AES, Groestl, on an ASIC chip, and use a low fpga with some SRAM for the main loop and the scratchpad.

If Bitmain already produce the sophon sc1 for AI, why they couldn't produce a similar one for mining like above ?

full member
Activity: 420
Merit: 184
I suspect - but admittedly don't know for sure - that the threat from FPGAs has been wildly overblown. FPGAs are extremely expensive compared to ASICs on a cost per computational effort basis, and so can't meaningfully compete with them. Take a look at this thread: https://bitcointalksearch.org/topic/diy-fpga-mining-rig-for-any-algorithm-with-fast-roi-3459858 which uses a VCU1525 that costs anywhere from $4000 to $8000 and can do about 22kH/s on CN v7.

Not really a threat to GPUs, even.

Are you aware that Bitmain already sold FPGA ? You should check about sophon sc1 ($500).  If they did in the past, why they couldn't produce others like a custom version of VCU1525 in the near future ?

VCU1525 is expensive because is a development board for prototyping ASIC before production, but they are less expensive ones and modded for crypto-mining : bcu1525 sell by OhGoodAGirl. ($3600)

No, I wasn't aware that Bitmain sold FPGAs, but I'm an EE by profession and have some (tangential) experience with FPGAs and trust me, they are always many times more expensive for a given amount of computational power than an ASIC; their sole advantage is that they can be reprogrammed and so the cost to get them doing something useful is much, much less than an ASIC. Basically, FPGAs are great for prototyping ASICs and for low-volume products that require real-time processing power (something which is hard to come by in a MCU or DSP).

The reasons for the cost disadvantage are that FPGAs consist of lots of small computational blocks (usually implemented with programmable "look-up tables" or LUTs) that are interconnected by a programmable matrix. The LUTs are capable of performing any Boolean operation so they tend to take up a lot more space than, say, a dedicated logic gate. In additional, laying down so much metallization for the interconnect matrix also greatly restricts the ultimate density of computational power per unit area (for the same reason you can't fit as many houses into a given area if all the streets are 8-lane highways instead of ordinary residential roads).

So a $500 FPGA probably has about the same computational power as a $5 ASIC; but you need to make 1000's to 10000's of the ASIC to justify the non-recurring engineering (NRE) cost, and, of course, it's a one-trick pony, so any change to its functionality requires creating a whole new IC mask, sending it off to the fab, waiting to get onto the fab's production schedule, etc... Whereas for an FPGA you just upload a new bitstream (as long as there are sufficient LUTs, memory and interconnects available to perform the task).

jr. member
Activity: 111
Merit: 1
So you suggest we should change the PoW and hard fork every week to fight against FPGA/ASIC because you can't buy one in your local store ?
Are you aware you are comparing between a cryptocurrency and a platform ?
How dApps are supposed to work if the platform is unstable or worse with security flaw ?


I suspect - but admittedly don't know for sure - that the threat from FPGAs has been wildly overblown. FPGAs are extremely expensive compared to ASICs on a cost per computational effort basis, and so can't meaningfully compete with them. Take a look at this thread: https://bitcointalksearch.org/topic/diy-fpga-mining-rig-for-any-algorithm-with-fast-roi-3459858 which uses a VCU1525 that costs anywhere from $4000 to $8000 and can do about 22kH/s on CN v7.

Not really a threat to GPUs, even.



Are you aware that Bitmain already sold FPGA ? You should check about sophon sc1 ($500).  If they did in the past, why they couldn't produce others like a custom version of VCU1525 in the near future ?

VCU1525 is expensive because is a development board for prototyping ASIC before production, but they are less expensive ones and modded for crypto-mining : bcu1525 sell by OhGoodAGirl. ($3600)
full member
Activity: 420
Merit: 184
So you suggest we should change the PoW and hard fork every week to fight against FPGA/ASIC because you can't buy one in your local store ?
Are you aware you are comparing between a cryptocurrency and a platform ?
How dApps are supposed to work if the platform is unstable or worse with security flaw ?


I suspect - but admittedly don't know for sure - that the threat from FPGAs has been wildly overblown. FPGAs are extremely expensive compared to ASICs on a cost per computational effort basis, and so can't meaningfully compete with them. Take a look at this thread: https://bitcointalksearch.org/topic/diy-fpga-mining-rig-for-any-algorithm-with-fast-roi-3459858 which uses a VCU1525 that costs anywhere from $4000 to $8000 and can do about 22kH/s on CN v7.

Not really a threat to GPUs, even.

newbie
Activity: 12
Merit: 0
Quote
v1 blocks cannot be mined (these are imported  blocks), rejecting
full member
Activity: 420
Merit: 184
...
will joining soon this project, and i believe this project will success in the future

Lol, first there's an open rebellion against the about-face on ASICs and now here come the professional shitposting sig-spammers.

Meanwhile, price has declined >40% since June 21. Can I get a golf clap anyone?

jr. member
Activity: 111
Merit: 1

Have fun with FPGAs and private ASIC miners.

And Dero isn't a fork anymore, it's an enhanced version of CryptoNote.

But I think you should listen the guy who spend $800k on a watch, after all it's more safe.

 Thanks buddy.

 Because in no way the exact same thing happens to any commercially available ASIC... oh no, it doesnt.
 And they would never refurb and sell soon-to-be doorstops out to the world after mining to kingdome come - they wouldnt dare.

 Fact is, GPU are democratic, ASIC are not.

 A 16 year old can build a miner in a week. It take substantial disposable income to risk pre-orders, group-buys or china imports.
 These are always done in bulk, even knowingly buying thousands of ASIC to mine for a few months and dump on ebays etc.
 Extremly short and limited warranties.

 ASIC, until mass produced by numerous world-wide OEM's, are a very closed-doors of insider priviledged.

 Big ASIC farms simply cannot speculate of future returns. First need to ROI as fast as humanly possible, then pay bill, then ... there is nothing.
 GPU's you have tons of re-sale value. Extended warranties. Buy it in every town, in every country, every nation - without restrictions.

 FPGA are reporgramable. ASICs are not.

 ASIC are only making chinese ASIC producers filthy rich. Nothing more, nothing less - good on them for being pioneers!

 GPU are much more flexible to switch coins, maintaining a far less negative sell-pressure on any give network. ASIC create insanely negative price pressure. Especially so on young coins. To the point where everyone looses interest.

 So yeah ... good luck DERO -  the new SUMO

So you suggest we should change the PoW and hard fork every week to fight against FPGA/ASIC because you can't buy one in your local store ?
Are you aware you are comparing between a cryptocurrency and a platform ?
How dApps are supposed to work if the platform is unstable or worse with security flaw ?
member
Activity: 294
Merit: 10
really nice project, DERO is a new, experimental blockchain technology written in Golang with a focus on enhanced Privacy and Smart Contracts while maintaining the transparency and security of the blockchain. The goal is to create a unique state of the art blockchain technology with enhanced Reliability, Privacy, Security, Usability, and Portability by bringing together some of the best proven technologies like CryptoNote Protocol and Smart Contracts, thereby allowing for the creation of Private Smart Contracts.
will joining soon this project, and i believe this project will success in the future
full member
Activity: 420
Merit: 184
Last blocks sync in progress, new wallet created.
But how we mine this now ?

With a Bitmain X3.
legendary
Activity: 1092
Merit: 1000

Have fun with FPGAs and private ASIC miners.

And Dero isn't a fork anymore, it's an enhanced version of CryptoNote.

But I think you should listen the guy who spend $800k on a watch, after all it's more safe.

 Thanks buddy, but learn your tech - it enhances nothing. It is never the best tech that wins, it is always the strongest community support that wins - this is the plain cold truth of a techological society. And this tech is not proven yet - there is no usecase, no real MVP, no product yet.

 Because in no way the exact same thing happens to any commercially available ASIC... oh no, it doesnt.
 And they would never refurb and sell soon-to-be doorstops out to the world after mining to kingdome come - they wouldnt dare.

 Fact is, GPU are democratic, ASIC are not.

 A 16 year old can build a miner in a week. It take substantial disposable income to risk pre-orders, group-buys or china imports.
 These are always done in bulk, even knowingly buying thousands of ASIC to mine for a few months and dump on ebays etc.
 Extremly short and limited warranties.

 ASIC, until mass produced by numerous world-wide OEM's, are a very closed-doors of insider priviledged.

 Big ASIC farms simply cannot speculate on future returns. First they need to ROI as fast as humanly possible, then pay for bills, then ... there is nothing left as new ASIC roll out to cannibalise earnings.

 GPU's you have tons of re-sale value. Extended warranties. Buy it in every town and every nation - without restrictions.

 FPGA are reporgramable. ASICs are not.

 ASIC are only making chinese ASIC producers filthy rich. Nothing more, nothing less - good on them for being pioneers!

 GPU are much more flexible to switch coins, maintaining a far less negative sell-pressure on any given network. ASIC create insanely negative price pressure. Especially so on young coins. To the point where everyone looses interest.

 When the largest supporting marketcap algo switches, there is zero incentive for ASIC producers to create a better version for tiny puny coins that are inherently worthless due to being so young. Soon, these ASIC will fry out, and no replacement. Back to GPU it will have to be.

 So yeah ... good luck DERO -  the new SUMO
sr. member
Activity: 661
Merit: 250
Last blocks sync in progress, new wallet created.
But how we mine this now ?
Pages:
Jump to: