Bitcoin Forum

I wonder if any of you tested the fury X with various mining algorithms?

AMD Radeon R9 Fury X Results:
– X11 default: 6.778 MHS
– X11 Wolf0 Mod: 8.123 MHS
– X13 default: 5.614 MHS
– X13 Wolf0 Mod: 7.176 MHS
– X15 default: 4.69 MHS
– X15 Wolf0 Mod: 6.335 MHS
– Quark modified: 22.37 MHS
– Qubit modified: 21.15 MHS
– Neoscrypt default: 147 KHS
– Lyra2RE default: 287 KHS
– Lyra2RE Pallas Mod: 450 KHS

Test from cryptomining blog

Too bad they're still not testing against (ocl)vanitygen .. I mean, if it runs at all, of course :)

Probably best if you take this to an AltCoin centric forum, rather than a Bitcoin centric forum.

Can this be real? the differences are huges. Not only a couple percent. the difference from 450KHS to 22.37MHS are enourmous. I did not think that they are so different...

yeah altcoins, they all looks the same  :-\

Why is Scrypt performance not on that list? 

Doesn't necessarily matter that GPUs aren't good for scrypt anymore, but moreso because scrypt performance is/was a pretty good baseline indicator of hashing chops between various cards.

280 could do ~600-700khs (unless you had one of those Gigabyte WF3 supercards :P )
280x could do 710-760khs
290/x could do 800-1000khs

Very good point. I hoped for some scrypt numbers too.

Well, as most know, every driver above 13.11 or so will give Hardware errors or bad hashrate when mining Scrypt.

13.11 drivers just simply do not work with Fury X cards.

I have 3 Fury's running and managed to get scrypt running without Hardware errors.

Here are the results.

I can see that the cards are not even doing their best to perform. If i run any other Algo the temps will go up to 45-50 Degrees.
I tried changing worksize, intensity, threads, with every change i did i got Hardware errors, bad results or even Freezing the system.

Results
http://s30.postimg.org/fs3zbmtmp/scrypt.png
http://s30.postimg.org/fs3zbmtmp/scrypt.png


Greetings!

I did some new tests.

New X11 result: 11Mhz
Screenshot
http://s2.postimg.org/5do903dsp/Schermafbeelding_2015_07_14_om_21_18_57.png

Will do some other tests later.

Greetings

The temps didn't go higher than what you see in that screenshot.

I let it run for 5 minutes or so, no issues, don't think there will be any..

Got 13mhz with worksize 128 but an occasional hw error after a few minutes.

Greetings

33 = low, i'm 100% sure that they can go higher, when i run my darkcoin mod on a 290x it goes to 7.7 Mhz, knowing that Wolf his mod easiliy reaches 9Mhz that means that my fury's should in theory easily touch 14Mhz..

The liquid cooling is indeed very awesome. they never exceed +45Degrees and that is when i'm really pushing them :)

It's around 25 Degrees in my warehouse. 20 cards in total,  all of them under 70Degrees, (i keep em that way though..)

So yeah, the Fury's are at full power only 10-15 Degrees hotter then my space :)

Nope. You are not right. The graphic card will dissipate the same amount of heat. Liquid or air cooling doesn't matter in such context. If the graphic card produces let's say 250W of heat it will dissipate 250W in your room anyway.

Quiete good result for x11, What about Quark algo? And also which kernel and miner are you using?is tyhat open source or private? İ 'll buyt some new cards, i didn't decide to buy fury x or 970's .Your answer so important to choosing cards :)))

On the furyx quark is doing 22 MHASH with the russian closed source miner.(5MHASH with the opensource miner)

2x 970 cards will cost around same as 1 furyx, and the hashrate is much higher. Lyra2re: 3.950 MHASH vs 420khash (9,4 times faster on the 970's)

True _sp, quark is only doing 23Mh. But this is with a non Fiji optimized binary.
If with the same source a fiji binary is compiled the hashrate will definitely hit 30 if not 35Mhz.

Regarding X11: with my mod i hit 11Mhz+, with the same mod i only get 7Mhz on a 290x. Knowing that i hit 9.5 mhz with wolf his mod/binary.
So knowing this, Fiji (fury x) will go much higher if optimized correct.

I'm waiting for Wolf to get his Fury X, then we will talk again :)


@fenomenhaa
It's my own X11 mod, but i'll release the binary if asked.

Greetings

Thanks M8's :))))

@Eliovp We discussed about just over hashing speed, There is also Power consumption side too.Fury X sure needs less energy. Did you checked the power consumption from wall?

The furyx reference design has a tdp of 275W and the gtx 970 has a tdp of 150W.

Interesting.

here are the hashrates for the gtx 980,gtx980ti and the titanx (reference cards, on standard clocks)

Tested with ccMiner 1.5.51 SP-MOD

The latest version(1.5.55) will give higher rates on lyra2,x11,quark etc

http://cryptomining-blog.com/wp-content/uploads/2015/06/nvidia-gtx-980-ti-hashrates.jpg

More at:

http://cryptomining-blog.com/4861-crypto-mining-performance-of-the-new-nvidia-geforce-gtx-980-ti/

@sp

TDP isn't power consumption.İt's Thermal Design Power,"It is maximum amount of heat generated by the CPU,GPU" And u know better then us,you are coder, power consumption depends on kernel efficeincy too(,so we caannot tkae TDP base of real Power consumption.) miners like us  :D just focused to power consumption from wall ))).I think fury X have less power consumpiton then dual 970's (i bios edited all of my 970's with %40 Tdp more .It fluctates 150-190W per card with o/c,avarege 180W+) I knew this diagram of maxwells's M8 :))

TDP is usually accurate for reference design card in terms of power consumption.  For example a reference design GTX970 will consume around 145W under full load while GTX980 will consume around 165W.
However, if the card is not based on reference design, then usually the manufacturer will increase the TDP to hit higher clock speed to squeeze out more performance.

The main problem is hashing rarely causes cards to go full load. I'm pretty sure groestl at 30.6MHs on 980Ti isn't full load nor qubit at 17.4.

The numbers seems too low, but these are reference cards.

My gigabyte 970 oc is doing 16.2 MHASH in the quark algo on the factory clock. Faster than the 980 reference in the cryptomining blog.
my zotac 980 oc card is doing 18 MHASH++

Running quark algo pushed my gtx980 over 100% load.  However, x11 algo seems to be slightly less than that.

Nice to know, I believe we should ask ourself questions at this point as that's impossible.
You probably don't know, officially my house heater has 109% efficiency.

Well, at least from the power consumption perspective it's hitting full load.  My card registered 200W when running quark algo.  From benchmark (of the same card) that I have read on other site, full load for my card is around 200W.  So, I think it is pretty close - minus the VRAM loading perhaps.

Ok, i finally did some power usage tests.

All these tests are with 3x R9 XFX Fury X, Intel I5, 16GB Corsair Vengeance, 2 x Corsair AX 860i

Test number 1:
This is an X11 test, all 3 cards doing 11.5 Mhz @ around 35 degrees.
Complete rig is pulling just over 680 Watt & 3.1 amps
Picture
http://s10.postimg.org/rjex7v0mh/IMG_7064.png


Test number 2:
Quark @ 23 Mhz per card @ around 38 Degrees.
Rig is pulling 776 Watt & 3.5 amps
Picture
http://s8.postimg.org/ingyj319h/IMG_7065.png


Test number 3:
Siacoin @ 1.9 Ghz per card @ around 40 Degrees.
Rig is pulling around 1006 Watt & 4.6 amps.
Picture
http://s13.postimg.org/uzpna2a5z/IMG_7066.png


Knowing all of this i can only conclude that there is much to do for X11 & Quark. I have a feeling X11 will definitely be able to go over 20Mhz and Quark should be able to go over 30Mhz with ease.
Also it is clear and allready known that they love power, as do almost all of the AMD cards.

in Case number 1, the complete rig is pulling almost 700 Watt, take 100 watt out (Processor, mobo and such) and a single card is @ around 200Watt, not bad though for doing 11+Mhz.

However as you can see in case number 3, the complete rig is pulling over 1000 Watt with only 3 cards. So they're easily pulling 300Watt per card. And i'm sure i haven't hit the top speed on Siacoin.


If there's anyone who wants to see a test on another algo. Just ask. No problem!


Greetings!

Siacoin hashrate is impressive! Are you using the public miner?
I'm interested in knowing hashrate and power usage for the groestlcoin/diamond algo.
Please use both the opensource v1 kernel and the experimental v2 binary in my thread.

Many thanks!

Hey Pallas.

no i'm not using the public miner, hashrate on the public miner is just above 1Ghz

As for your Groestl/diamond interest.

I installed this version of sgminer
https://github.com/prettyhatemachine/sph-sgminer

when running Groestl/diamond on stock settings "1050 engine" i hit around @ 778 Watt
http://s27.postimg.org/ja3i6yjnn/groestl_Stock.png
Watt
http://postimg.org/image/v14qbrulp/


When i use your Hawaii binary, rename it, run it, i run around 40Mhz+ but with HW errors
http://s17.postimg.org/yt38buk2n/groestl_Pallas.png
Hashrate was still rising and this is with stock clocks. So with 1100 (which those cards can easily handle it should go to 50Mhz)



Oh and a screenshot of Siacoin speed

http://s15.postimg.org/ropza5kh7/sia.png

Greetings

thanks for the tests.
in the end the performance is 0-20% higher than a 290x depending on the algo:

sia - about the same
groestl - about 10% more
x11 and quark - about 20% more

let's hope they make a better opencl compiler to unleash its full power!
finally a curiosity: did you get valid blocks with that ultra fast sia miner?

No problem :)


Mm, Sia about the same? on my 290x i hit around 1.25 Ghz (oc'ed), on my Fury's i hit 1.9 Ghz (stock clocks)
That's a difference of 40%, non open source though..

As you said, a better opencl compiler and optimized kernels would be seriously interesting. Very curious what they are capable of.

Yes i have found plenty of blocks with that miner :)


Greets

Ok, got some new X13 results!

Also own Mod

9.7Mhz per card @ 1050 Engine clock pulling 756Watt with 3 cards.

http://s13.postimg.org/umgflbgc7/marucoin_Ok.png

Watt

http://s29.postimg.org/ksobyeh93/IMG_7069.png


Greetings!

anyone tried to mine XMR with the Fury X?

@Eliovp Thanks for your detailed power consumption and hashrate charts.Fury X really interesisting card, pls keep updating informations,Still can't deceide to buy 970 or fury X :)))

Ok here are some details.

Around 1300 h/s per card

screenshot
http://s24.postimg.org/fve5l9zhx/Schermafbeelding_2015_07_24_om_21_34_24.png

Power usage at that time
http://s3.postimg.org/y49u7m02r/IMG_7071.png


Greetings

would be nice if the images was a tad smaller

Would be even nicer if your resolution was a tad bigger :-)

I'll remove the image tags, specially for you :-)

Thank you for testing.  :)

Is the 490w for all 3 cards combined?

Ok i have some new tests regarding Cryptonote.

As seen in the screenshot. 1 Fury at stock settings runs @ around 850+h/s

Cards were almost at idle temperatures. So in other words, they were hardly doing their best :)

Screenshot of hash rate
http://s29.postimg.org/k8yc7nivr/claymore.png

As i said before, with those low temps, i know this also could go a lot faster. Energy usage confirmed that.
Not even 500 Watt with 3 Cards.

Screenshot
http://s9.postimg.org/anraaxfyn/IMG_7077.png


I have only one algo that drives my fury's over 40 degrees. And that is Siacoin. This is because i'm pretty sure that they're almost working at full power on that algo. Also the power usage when running sia is more then double than what it was pulling on Cryptonote. Not only Cryptonote but also X11, Quark etc..


Again, if someone wants to test something, Just ask :)


Greetings!

Thanks for posting! This definitely helps inform my next purchase. I've been debating going the AMD route for a new monero rig. So at current release, it seems that for whatever reason the fury performs as well as the 290x.

>As i said before, with those low temps, i know this also could go a lot faster. Energy usage confirmed that.

The bottleneck of CryptoNight PoW algo is that it requires loading small data blocks from random memory addresses, a lot of times. Because addresses are random, memory cache is useless. You see low temps because GPU core does not work all the time, most time it waits data from GPU memory due its latency. And it cannot be optimized because the algo was designed to use memory latency to be asic-resistant. Fury X has 4096 bit memory bandwidth, but it is useless for this algo because big bandwidth is not related to memory latency.

same for lyra2re (it looks like fury x performance is worse than 280x), but they are changing the algo (at least for vertcoin) in order to be more gpu-friendly, so that might change.

You're very welcome!

It's useless for most algo's..
But again, everything was at stock settings, i have been playing around with the core engine and was able to push it to 1190 without a problem. After that it started to fail, temperatures were still very low though, if i was on windows i could play with voltage as it's possible with Afterburner.

I'm not experienced enough to optimize any algo to be Fury X friendly :) But i know someone who is and will be playing with it in the next weeks.
I'm curious about the results!


Lyra2re performance is worse than a 280x you mean? i haven't tested that algo "yet".. so i don't know. I ignored Cryptomining blog results because most of them aren't correct compared to my results.

Regarding Cryptonote, seeing the results with other cards (stock settings) the fury beats them all, not by much, but it's still the fastest it seems :)


Greetings

Did a quick Lyra2re test with stock settings and open source kernel.

Dunno where that cryptoblog got their results.. but they're way off..

here are my results.

http://s21.postimg.org/rpe511cuf/Lyra2_RE.png

Almost 1mhz. That's a serious difference than the 430Kh Cryptoblog wrote...

Energy usage

http://s23.postimg.org/whnyc8f63/IMG_7080.png


@Pallas, tried to run your binary but that froze up my rig each time.
it doesn't freeze up when doing that for Quark or any other algo.

Greetings

It is opensource now, you don't need to use the binary...

Oh lol, didn't know that :) will do some more testing later then :)

I keep seeing people talk about "non-opens source" mining software.  Where can we purchase it?  I have tried to find it, but all I have found is open source software.

Google wolf0. He has made a neoscrypt kernal that is 300% faster than the opensource kernal. I am not sure he will sell it though because some of the kernals have been stolen and distributed on the internet for free.
With wolf0's code, a single 290x will outperform 2 furyx mining neoscrypt. That's why sharks with 1000+ gpu's are willing to pay alot of BTC to keep the kernals private.

So is there any private miner that can boost neoscrypt to >200% on Nvidia? :P ;D

Good point.  It would be very difficult, if not impossible, to make enough money selling it, before some jerk decided it needed to be free to everyone, and post it forums for everyone to have.  Being a programmer, I completely sympathize with that.  And I don't have a fat wad of cash sitting around to pay for something like that.  It is looking more and more like I am going to have to add mining software development to my to-do list.

Man I didn't even know AMD released new GPUs. However looking at the prices, it seems that it would take around 600 days or so to break-even on these GPUs alone (not including mobo, PSUs, and electrcity costs), im surprised people are still buying GPUs for mining.

you must take resell price into account.
some cards can be sold at 90% buy price.

Did some tests on Quark this time.

hitting a stable 23.5 Mhz @ 820 Watt(3 fury's)
Very sure they can still go higher.


Big thx to Wolf0 for helping out :)!

Screenshot

http://s18.postimg.org/4iyau1ne1/quark_Fury.png

offtopic: wolf, something reminded me of you last weekend.   :)
https://i.imgur.com/qETDOgl.jpg

How fast is the opensource miner?

everyone forget quark algo, big hashs made from FPGE miner. i shutdown miner today and holiday fews day. profits very low now.  :D

What is a FPGE miner?  ;D '

Sharkcoin is taking a dip, and the profitability is down.

http://coinmarketcap.com/currencies/sharkcoin/

like integrated circuit. i know they are running few months(single chip using around 60watts ）. i cannot talking more information, sorry. bcz it's business to someone. i hold a lots of GPU miners in my space, i working building project after few month then i will coming back to mining, and consider mining new aglo coins after.

maybe you mean FPGA :D

yes...

I dont see a spike in quark (or sharkcoin) hashrate.

Quark profits are down cause SAK price dipped, not because of new tech (FPGA) or improved miner.

and the algo for FPGA??

whew! that leaves monero out of your sights then! Godspeed!

 
can make dual miner XMR with ETH maybe?