elective-stereophonic
elective-stereophonic
[ANN] Jinn
Please login or register.

Login with username, password and session length
Advanced search  

News:

Latest Nxt Client: Nxt 1.11.15

Pages: 1 ... 18 19 [20] 21 22 ... 80

Author Topic: [ANN] Jinn  (Read 323529 times)

Sebastien256

  • Hero Member
  • *****
  • Karma: +169/-24
  • Offline Offline
  • Posts: 2823
  • ^LOOK UP^ = Nxt community!
    • View Profile
Re: [ANN] Jinn
« Reply #380 on: September 27, 2014, 07:12:38 pm »

median? first bid on 50000 assets is at 100Nxt
70Nxt is bottom at 100000 assets.
Logged
Please drop your ideas concerning Nxt and/or NRS in this topic -> List of feature request for Nxt and/or NRS (with the full list in OP).

rtrtcrypto

  • Sr. Member
  • ****
  • Karma: +37/-5
  • Offline Offline
  • Posts: 279
    • View Profile
Re: [ANN] Jinn
« Reply #381 on: September 27, 2014, 07:14:06 pm »

Time to accept our money  ;D

Logged

Sebastien256

  • Hero Member
  • *****
  • Karma: +169/-24
  • Offline Offline
  • Posts: 2823
  • ^LOOK UP^ = Nxt community!
    • View Profile
Re: [ANN] Jinn
« Reply #382 on: September 27, 2014, 07:14:48 pm »

Time to accept our money  ;D

yes, this is kinda expensive now, imo.
Logged
Please drop your ideas concerning Nxt and/or NRS in this topic -> List of feature request for Nxt and/or NRS (with the full list in OP).

cobaltskky

  • Hero Member
  • *****
  • Karma: +226/-138
  • Offline Offline
  • Posts: 2344
    • View Profile
Re: [ANN] Jinn
« Reply #383 on: September 27, 2014, 07:16:53 pm »

Time to accept our money  ;D

yes, this is kinda expensive now, imo.

A greed.

SEE WHAT I DID THERE? lol
Logged

Sebastien256

  • Hero Member
  • *****
  • Karma: +169/-24
  • Offline Offline
  • Posts: 2823
  • ^LOOK UP^ = Nxt community!
    • View Profile
Re: [ANN] Jinn
« Reply #384 on: September 27, 2014, 07:22:29 pm »

Time to accept our money  ;D

yes, this is kinda expensive now, imo.

A greed.

SEE WHAT I DID THERE? lol

To give an idea, for 70 Nxt, Jinn is approximatly valued as high as superNet at the moment (end of spuernet ICO). This make Jinn a very big crowdfunding...
« Last Edit: September 27, 2014, 07:24:34 pm by Sebastien256 »
Logged
Please drop your ideas concerning Nxt and/or NRS in this topic -> List of feature request for Nxt and/or NRS (with the full list in OP).

cobaltskky

  • Hero Member
  • *****
  • Karma: +226/-138
  • Offline Offline
  • Posts: 2344
    • View Profile
Re: [ANN] Jinn
« Reply #385 on: September 27, 2014, 07:24:07 pm »

Time to accept our money  ;D

yes, this is kinda expensive now, imo.

A greed.

SEE WHAT I DID THERE? lol

To give an idea, for 70 Nxt, Jinn is approximatly valued as high as superNet at the moment (end of spuernet ICO).

With no working assets?  SuperNET has MGW at least...I believe the "onioning" thingy that James is attempting to accomplish had a success and public test this past week.
Logged

TwinWinNerD

  • Hero Member
  • *****
  • Karma: +222/-116
  • Offline Offline
  • Posts: 2012
  • CEO BitPanda.com
    • View Profile
Re: [ANN] Jinn
« Reply #386 on: September 27, 2014, 07:57:21 pm »

Time to accept our money  ;D

yes, this is kinda expensive now, imo.

A greed.

SEE WHAT I DID THERE? lol

To give an idea, for 70 Nxt, Jinn is approximatly valued as high as superNet at the moment (end of spuernet ICO).

With no working assets?  SuperNET has MGW at least...I believe the "onioning" thingy that James is attempting to accomplish had a success and public test this past week.

It has inventory, that alone is worth the marketcap it is traded at.

Zahlen

  • Full Member
  • ***
  • Karma: +26/-4
  • Offline Offline
  • Posts: 228
    • View Profile
Re: [ANN] Jinn
« Reply #387 on: September 27, 2014, 09:12:33 pm »

This is really interesting! Some questions:

1) The FAQ mentions that ternary "gives the highest density of information representation among other integer bases". (Base 5 would be even denser than base 3, but yeah it's still denser than binary.) Will Triangle also be developing ternary storage, or will Jinn be interfacing with conventional binary storage?

2) How will you go about making the prototype Jinn? Both the circuit design and the fabrication. This seems to me the most difficult part of your plan. This breaks down into sub-questions like Who do you have in mind to design the circuits and do the fabrication? What experience do they have with making true-false-unknown logic-based circuits? And making horizontally scalable chips? And cost/resource use efficiency? (Not sure if I'm asking the right questions here, I dunno much about the prototyping process but would like to learn more. Maybe someone can refine the questions.)

3) How will you go about testing the Jinn chips? Both the prototypes and the manufactured versions.

4) If, at some point during development, funds raised from previous steps run out, what are your plans for raising additional funds?
« Last Edit: September 27, 2014, 09:14:47 pm by Zahlen »
Logged

Come-from-Beyond

  • Hero Member
  • *****
  • Karma: +794/-671
  • Offline Offline
  • Posts: 4013
    • View Profile
Re: [ANN] Jinn
« Reply #388 on: September 27, 2014, 09:35:59 pm »

This is really interesting! Some questions:

1) The FAQ mentions that ternary "gives the highest density of information representation among other integer bases". (Base 5 would be even denser than base 3, but yeah it's still denser than binary.) Will Triangle also be developing ternary storage, or will Jinn be interfacing with conventional binary storage?

2) How will you go about making the prototype Jinn? Both the circuit design and the fabrication. This seems to me the most difficult part of your plan. This breaks down into sub-questions like Who do you have in mind to design the circuits and do the fabrication? What experience do they have with making true-false-unknown logic-based circuits? And making horizontally scalable chips? And cost/resource use efficiency? (Not sure if I'm asking the right questions here, I dunno much about the prototyping process but would like to learn more. Maybe someone can refine the questions.)

3) How will you go about testing the Jinn chips? Both the prototypes and the manufactured versions.

4) If, at some point during development, funds raised from previous steps run out, what are your plans for raising additional funds?

1)
Quote
In a numerical system,
the number N is given by N= Rd where R is the
radix and d is the necessary number of digits
up to the next highest integer value where
necessary. If the cost or complexity C in any
system is assumed to be proportional to R x D, then
C= k(R x d) =k[R (ln N/In R)] (2)
where k is some constant. Differentiating with
respect to R will show that for a minimum cost
C, R should be equal to e (2.718).Since in
practice R must be an integer, this suggests
that R=3(ternary) would be more economical
than R =2(binary)

2) We will hire specialists to solve this issue.

3) The same as 2)

4) It's not a technical question, I can't answer it
Logged

Zahlen

  • Full Member
  • ***
  • Karma: +26/-4
  • Offline Offline
  • Posts: 228
    • View Profile
Re: [ANN] Jinn
« Reply #389 on: September 27, 2014, 09:43:38 pm »

Ah, I get what you mean by information 'density' now, thanks. So Jinn will still be working with binary storage then (and performing conversions as needed)?

(The 'suggests' step can be made rigourous by simply calculating 3 / log 3 < 2 / log 2, and showing than n / log n is a increasing function from 3 onwards).

Do you have any specialists in mind for 2) and 3)?
« Last Edit: September 27, 2014, 09:46:52 pm by Zahlen »
Logged

Windjc

  • Sr. Member
  • ****
  • Karma: +59/-17
  • Offline Offline
  • Posts: 360
    • View Profile
Re: [ANN] Jinn
« Reply #390 on: September 27, 2014, 09:50:45 pm »

Time to accept our money  ;D

yes, this is kinda expensive now, imo.

I had to bow out at 61.1

Lets see if that was a smart move.

I don't see how this won't go through a period of dumpage at some point. But maybe I'm wrong about that. In fact, I hope I am, as that will mean some crazy awesome stuff is being developed and quickly.
Logged

Come-from-Beyond

  • Hero Member
  • *****
  • Karma: +794/-671
  • Offline Offline
  • Posts: 4013
    • View Profile
Re: [ANN] Jinn
« Reply #391 on: September 28, 2014, 06:34:05 am »

Do you have any specialists in mind for 2) and 3)?

If you are asking personally me, then no.
Logged

tylergillies

  • Jr. Member
  • **
  • Karma: +4/-0
  • Offline Offline
  • Posts: 42
    • View Profile
Re: [ANN] Jinn
« Reply #392 on: September 28, 2014, 06:39:13 am »

Those who say that Jinn is way overpriced now and the most risky asset, I don't agree. Look at what SuperNET did as far as funding goes. It's no less riskier than Jinn. Either both of them are risky and overpriced, or neither of them are. I guess we have to wait for a few months to find out.

Really?
Supernet is backed by assets.
Jinn is not.
You canĀ“t compare them.

To value this project at 2,5mio$ is insane. I know they need funding. They really do, but 60 NXT per share is insane.

I wouldn't really call those assets. To me both SuperNET and Jinn are about the same high risk, as both projects essentially rely on one person's vision (James for SuperNET, CfB for Jinn) with a lot of spots that investors don't comprehend, not to mention other risks. But you are free to have your own view on them. I'll shut up now or someone will say I am spreading FUD :)

SuperNet is not just backed by James vision. Money talks. Bullshit walks. I invest in James because he has proven to make money hand over fist on multiple assets.
Logged

blackyblack1

  • Hero Member
  • *****
  • Karma: +165/-82
  • Offline Offline
  • Posts: 1763
    • View Profile
Re: [ANN] Jinn
« Reply #393 on: September 28, 2014, 07:05:49 am »

This is really interesting! Some questions:

1) The FAQ mentions that ternary "gives the highest density of information representation among other integer bases". (Base 5 would be even denser than base 3, but yeah it's still denser than binary.) Will Triangle also be developing ternary storage, or will Jinn be interfacing with conventional binary storage?

2) How will you go about making the prototype Jinn? Both the circuit design and the fabrication. This seems to me the most difficult part of your plan. This breaks down into sub-questions like Who do you have in mind to design the circuits and do the fabrication? What experience do they have with making true-false-unknown logic-based circuits? And making horizontally scalable chips? And cost/resource use efficiency? (Not sure if I'm asking the right questions here, I dunno much about the prototyping process but would like to learn more. Maybe someone can refine the questions.)

3) How will you go about testing the Jinn chips? Both the prototypes and the manufactured versions.

4) If, at some point during development, funds raised from previous steps run out, what are your plans for raising additional funds?
There are some contracting ASIC manufacturers. Chip can be designed and emulated with Verilog-like tools. But ASIC manufacturing process is very expensive and takes a lot of time.
Logged

jabo38

  • Sr. Member
  • ****
  • Karma: +40/-38
  • Offline Offline
  • Posts: 381
    • View Profile
Re: [ANN] Jinn
« Reply #394 on: September 28, 2014, 07:58:26 am »

A high bidder for 5000 just cancelled his order.  Price remains at 70 but has 5000 more 70s to go through than a little while ago.   
Logged
Never Enough Money

Brangdon

  • Hero Member
  • *****
  • Karma: +229/-25
  • Offline Offline
  • Posts: 1389
  • Quality is addictive.
    • View Profile
Re: [ANN] Jinn
« Reply #395 on: September 28, 2014, 09:40:13 am »

Erlang-based solutions hit the wall after certain threshold. This seems to be a fundamental flaw of its paradigm, because several people from independent teams working in Belarusian gamedev confirm this phenomenon and search on the Internet doesn't give a good solution.
Is that because it isn't ternary, or is there more going on? How important is ternary to what you are doing? My intuition is that the increase in storage density would have a fairly marginal effect, and not worth the hassle. If you dropped the ternary thing, what difference would it make?
Logged

Zahlen

  • Full Member
  • ***
  • Karma: +26/-4
  • Offline Offline
  • Posts: 228
    • View Profile
Re: [ANN] Jinn
« Reply #396 on: September 28, 2014, 10:36:13 am »

Do you have any specialists in mind for 2) and 3)?

If you are asking personally me, then no.

There are some contracting ASIC manufacturers. Chip can be designed and emulated with Verilog-like tools. But ASIC manufacturing process is very expensive and takes a lot of time.

Thanks for the replies. But it sounds like Jinn will be much harder to design, manufacture and test than ASICs, or even GPUs. These are focused on fairly specific tasks like computing hashes or matrix multiplication, I reckon you can reuse (and incrementally improve on) old designs and testing methods. So I'm not sure most ASIC manufacturers will be up to the task.

There are actual ternary logic computing devices from decades ago, and there's a lot of research on ternary (and other) logics. So the expertise exists, but the practice may be dated. Also, current message passing approaches for multiple processors incur a lot of overheads and don't scale well, Jinn will need a different approach (maybe something like packet routing, I read something about this a while ago but don't have a link).

(I'm not speaking as someone in the industry, all this is from just Google and general computing knowledge, so maybe I'm wrong about some stuff.)


Right now this key step of the project feels too unknown and risky for me, so I'll have to sit out for now :) I'd feel a lot more confident about investing once you find/attract the expertise. If existing manufacturing and testing tools can't be easily adapted, it'll also take more time and $ to develop the toolchain.

But this is really, really cool! Best wishes to Triangle, and I'll keep checking in to the project from time to time.
Logged

valarmg

  • Hero Member
  • *****
  • Karma: +178/-57
  • Offline Offline
  • Posts: 1766
    • View Profile
Re: [ANN] Jinn
« Reply #397 on: September 28, 2014, 10:54:56 am »

Do you have any specialists in mind for 2) and 3)?

If you are asking personally me, then no.

There are some contracting ASIC manufacturers. Chip can be designed and emulated with Verilog-like tools. But ASIC manufacturing process is very expensive and takes a lot of time.

Thanks for the replies. But it sounds like Jinn will be much harder to design, manufacture and test than ASICs, or even GPUs. These are focused on fairly specific tasks like computing hashes or matrix multiplication, I reckon you can reuse (and incrementally improve on) old designs and testing methods. So I'm not sure most ASIC manufacturers will be up to the task.

There are actual ternary logic computing devices from decades ago, and there's a lot of research on ternary (and other) logics. So the expertise exists, but the practice may be dated. Also, current message passing approaches for multiple processors incur a lot of overheads and don't scale well, Jinn will need a different approach (maybe something like packet routing, I read something about this a while ago but don't have a link).

(I'm not speaking as someone in the industry, all this is from just Google and general computing knowledge, so maybe I'm wrong about some stuff.)

I worked in IC design for many years, and I think it's close to impossible for ternary logic devices to get near to competing with binary design within any kind of reasonable timeframe even with infinite resources.

But Come-from-Beyond mentioned running their ternary logic initially on binary-based ICs, and long term creating wholly ternary designs on optical chips which seems more possible (though I don't know much about optical device design.)
Logged
NXT-CSED-4PK5-AR4V-6UB5V

Come-from-Beyond

  • Hero Member
  • *****
  • Karma: +794/-671
  • Offline Offline
  • Posts: 4013
    • View Profile
Re: [ANN] Jinn
« Reply #398 on: September 28, 2014, 11:24:49 am »

Is that because it isn't ternary, or is there more going on? How important is ternary to what you are doing? My intuition is that the increase in storage density would have a fairly marginal effect, and not worth the hassle. If you dropped the ternary thing, what difference would it make?

I don't know what causes the problem in Erlang, ternary has nothing to do with it. Jinn can be binary and still be very valuable, but it doesn't make much sense to create new architecture on base of obsolete technology. Binary will become obsolete right after optical computers hit the market.
Logged

Come-from-Beyond

  • Hero Member
  • *****
  • Karma: +794/-671
  • Offline Offline
  • Posts: 4013
    • View Profile
Re: [ANN] Jinn
« Reply #399 on: September 28, 2014, 12:03:32 pm »

But it sounds like Jinn will be much harder to design, manufacture and test than ASICs, or even GPUs. These are focused on fairly specific tasks like computing hashes or matrix multiplication...

Jinn doesn't need to support legacy instruction sets and it will require much lower number of logic gates. After the instruction set is ready you will see that it's very simple (3 opcodes only, 1 trit is enough to encode it). I can't say how many elements will be required for a single core, but I suspect it will be near the same as Bitcoin ASICs have, because architecture of Jinn mimics fractals - simple components form sophisticated structures.
Logged
Pages: 1 ... 18 19 [20] 21 22 ... 80
 

elective-stereophonic
elective-stereophonic
assembly
assembly