Claymore's ZCash/BTG AMD GPU Miner v12.6 (Windows/Linux)

[Daniel0785]

Would you please do some further optimizations for 380? It's a bit too far away from 280.

[1714716421]

1714716421

[1714716421]

1714716421

[1714716421]

1714716421

[christiano88]

It is normal for the 380 to be far away from the 280, 380 has 256bit bus and 280 has 384bit, so theoretically it has 50% more bandwidth (it also comes down to the difference in frequency though).

@Claymore, what about Fury (non-X) cards? How much will they have with your new update?

[Taxidermista]

What Windows are you guys using? 7 or 10? Which one gives more H/s?

Win 8.1

For more than 5 cards in 1 mobo you need to install Win 10.

Hi

that is not correct ! I'am using Win 8.1 with 6 Cards ! Thats working with a driver patch !

Have a nice day

Sorry, you're right about that. I just experienced that Win 8.1 is not stable with 6 or 7 cards (using that patch), so usually I don't recommend Win 8.1 for larger rigs. (But maybe not the driver patch or Win 8.1 was the cause of the unstability.)

Windows 8 and Windows 7 has been perfectly stable with 6 cards without any patch, for YEARS.

[Claymore]

Memory bandwidth:

RX480: 256 GB/s
390X: 384 GB/s

ZEC uses a lot of memory operations. Do you still think that RX480 can work as fast as 390X?

[Claymore]

Hey man look some one is trying to make some coins... https://satoshibox.com/........

It's scam and/or virus, there is no v8, I don't have it yet.

[sorry2xs]

Memory bandwidth:

RX480: 256 GB/s
390X: 384 GB/s

ZEC uses a lot of memory operations. Do you still think that RX480 can work as fast as 390X?

my 390x claim to have a 512 gb bandwidth  have i been lied to by the manufactures  
Edit: i thoguth that the 7950"s have 384 gb bandwidth

[melloyellow]

Memory bandwidth:

RX480: 256 GB/s
390X: 384 GB/s

ZEC uses a lot of memory operations. Do you still think that RX480 can work as fast as 390X?

my 390x claim to have a 512 gb bandwidth  have i been lied to by the manufactures  
Edit: i thoguth that the 7950"s have 384 gb bandwidth

You're thinking of memory bus, 512 bit. 

[fittsy]

You can already do this by mining at an auto-switching pool using Claymore. What would be useful is for Claymore to auto-switch POOLS based on profit, he can dream big and combine all his miners controlled by a separate program that looks at the pool API's and starts the appropriate miner.

Can you imagine hundreds of thousands of miners hitting that API?

How do you think the nicehash miner works?

[fittsy]

my experience is as follows i bought 30 v.3 usb risers 10 of the are in current use another ten are in the recycling bin and the remainder are new for spares

I just bought a V3, solder is fucked up... Returning and buying the V6

Link to these risers?

My v6 risers that I've been selling for 3 months are linked in my sig.

I'm almost out but I have more on the way from China already.

[sorry2xs]

Memory bandwidth:

RX480: 256 GB/s
390X: 384 GB/s

ZEC uses a lot of memory operations. Do you still think that RX480 can work as fast as 390X?

my 390x claim to have a 512 gb bandwidth  have i been lied to by the manufactures  
Edit: i thoguth that the 7950"s have 384 gb bandwidth

You're thinking of memory bus, 512 bit.  

here i thought  memory bandwidth pertain to memory bus

[nitromining]

You can already do this by mining at an auto-switching pool using Claymore. What would be useful is for Claymore to auto-switch POOLS based on profit, he can dream big and combine all his miners controlled by a separate program that looks at the pool API's and starts the appropriate miner.

Can you imagine hundreds of thousands of miners hitting that API?

How do you think the nicehash miner works?

I did that with BAMT switching between scrypt, X11, and X15. "Back in the day"

[Mickspabtc]

hi am running 6 x rx 470 gb all standard hashing as 150/sols/s a card but hear people getting 170 a card is there anyway i can up this i have played with afterburner but seems to crash them alot...any modded bios for them? also hvae 2 rx 480 rigs hashing 170 but seen reports of 190-200??

[manimj]

Hey man look some one is trying to make some coins... https://satoshibox.com/ea83howjfpzbzzehdzd5a652

It's scam and/or virus, there is no v8, I don't have it yet.

Thats what i thought. Just want to keep you posted.
Thank you for your hard work & helping us all.

[ol92]

i feel 280x the cards Claymore loves the most - will have 200 sols per card i nthe update )

I like 390-390X the most - I'm going to reach 300H/s on stock clocks.
RX480 will show about 190-200 I think.
280X - about 200 or a bit more.

and what about nano/fury ? They have 512 gGB of bandwith...

[Claymore]

i feel 280x the cards Claymore loves the most - will have 200 sols per card i nthe update )

I like 390-390X the most - I'm going to reach 300H/s on stock clocks.
RX480 will show about 190-200 I think.
280X - about 200 or a bit more.

and what about nano/fury ? They have 512 gGB of bandwith...

Yes, but too wide memory bus, 4096bit is too much for most PoW algos and therefore cannot be used completely.
Nano will show about 250H/s, may be I will reach a bit more.

[Heatspeed]

i feel 280x the cards Claymore loves the most - will have 200 sols per card i nthe update )

I like 390-390X the most - I'm going to reach 300H/s on stock clocks.
RX480 will show about 190-200 I think.
280X - about 200 or a bit more.

What Hashrate will a R9 Nano have on v8?
Is it even possible for a rx480 to get a hashrate above 200 H/s?
I'm trying to figure out whats the best investment for another rig..
Btw. good work!

[Golku]

i feel 280x the cards Claymore loves the most - will have 200 sols per card i nthe update )

I like 390-390X the most - I'm going to reach 300H/s on stock clocks.
RX480 will show about 190-200 I think.
280X - about 200 or a bit more.

What Hashrate will a R9 Nano have on v8?
Is it even possible for a rx480 to get a hashrate above 200 H/s?
I'm trying to figure out whats the best investment for another rig..
Btw. good work!

Somebody analyses the hash rate of the RX 470. The limit is around the 170 H/s. So it will not be too high.

170 for a 470 how in the world are you going to achieve that without getting ridiculous high wattage?

Im hashing with my 470 at 150 with a decent wattage

[Mastker]

i feel 280x the cards Claymore loves the most - will have 200 sols per card i nthe update )

I like 390-390X the most - I'm going to reach 300H/s on stock clocks.
RX480 will show about 190-200 I think.
280X - about 200 or a bit more.

What Hashrate will a R9 Nano have on v8?
Is it even possible for a rx480 to get a hashrate above 200 H/s?
I'm trying to figure out whats the best investment for another rig..
Btw. good work!

Somebody analyses the hash rate of the RX 470. The limit is around the 170 H/s. So it will not be too high.

170 for a 470 how in the world are you going to achieve that without getting ridiculous high wattage?

Im hashing with my 470 at 150 with a decent wattage

nerdralph said

While my initial analysis was focused on the external GDDR5 bandwidth limits, current ZEC GPU mining software seems to be limited by the memory controller/core bus.  On AMD GCN, each memory controller can xfer 64 bytes (1 cache line) per clock.  In SA5, the ht_store function, in addition to adding to row counters, does 4 separate memory writes for most rounds (3 writes for the last couple rounds).  All of these writes are either 4 or 8 bytes, so much less than 64 bytes per clock are being transferred to the L2 cache.  A single thread (1 SIMD element) can transfer at most 16 bytes (dwordX4) in a single instruction.  This means a modified ht_store thread could update a row slot in 2 clocks.  If the update operation is split between 2 (or 4 or more) threads, one slot can be updated in one clock, since 2 threads can simultaneously write to different parts of the same 64-byte block.  This would mean each row update operation could be done in 2 GPU core clock cycles; one for the counter update, and one for updating the row slot.

Even with those changes, my calculations indicate that a ZEC miner would be limited by the core clock, according to a ratio of approximately 5:6.  In other words, when a Rx 470 has a memory clock of 1750Mhz, the core would need to be clocked at 1750 * 5/6 = 1458Mhz in order to achieve maximum performance.

If the row counters can be kept in LDS or GDS, the core:memory ratio required would be 1:2, thereby allowing full use of the external memory bandwidth.  There is 64KB of LDS per CU, and the AMD GCN architecture docs indicate the LDS can be globally addressed; i.e. one CU can access the LDS of another CU.  However the syntax of OpenCL does not permit the local memory of one work-group to be accessed by a different work-group.  There is only 64KB of GDS shared by all CUs, and even if the row counters could be stored in such a small amount of memory, OpenCL does not have any concept of GDS.

This likely means writing a top performance ZEC miner for AMD is the domain of someone who codes in GCN assembler.  Canis lupus?

[BenCodie]

can you guys share their speeds on stock setting of MSI R9 390 Gigabyte R9 390 Sapphire R9 390 and XFX R9 390 all 8GB
again stock setting no OC no modded bios please.

Thanks

[sorry2xs]

can you guys share their speeds on stock setting of MSI R9 390 Gigabyte R9 390 Sapphire R9 390 and XFX R9 390 all 8GB
again stock setting no OC no modded bios please.

Thanks

well i have msi r9 390x2 with the stupid lights and all they produce about 215/225 Hs each no nothing, and 3x r9 390 asus with stupid flashing lights performing about the same 215/225 each card for what is worth.