Of course it’s worth testing. Optimised or not (and I suspect that aspect of zcashd is in decent shape), it still allows for comparison of different systems.
2 Likes
Hey @voluntary, great work on setting up the wiki table for performance!
I’d recommend putting filters on the columns, so that we can order descending/ascending by equihash-10, for example. Wish I could help, but busy right now. Will take a look later if you haven’t already solved it! Cheers, tom
Hard to edit tables, here are mine if anyone is interested in adding them for reference
Intel i7-3930K (3.8 GHz)
16GB (4x4GB) DDR3 1600
25sec
procgenlimit=1
bare
i7-3630QM (3.4 GHz)
6GB DDR3 1600
33sec
procgenlimit=1
bare
Intel Xeon E5-2676 2.4 GHz (m4.large)
59.1 sec
procgenlimit=1
VM
Intel Xeon E5-2676 2.4 GHz (m4.large)
45.3 sec
procgenlimit=2
VM
2 Likes
@Shawn runs that site.
Cool, @Shawn, can you do this, or want me to take a look later? chrs
I’d recommend putting filters on the columns, so that we can order descending/ascending by equihash-10, for example. Wish I could help, but busy right now. Will take a look later if you haven’t already solved it! Cheers, tom
Please feel free to have a look and see what you can do. Wikimedia has code for the tables that is tricky to use so I have been having people use the table generator and copy/paste the results. Unfortunately that doesn’t really have advanced table functions like sort/filter.
If you have a better method I would l be happy to use it.
Thanks @trolloniex I will add yours and blackhatzw info to the table when I get some free time.
I stopped the VM in AWS yesterday and start it today but zcash server can’t start anymore, did you hit this issue before?
ubuntu@ip-xxxxxxxx:~/zcash/src$ ./zcashd -daemon
Zcash server starting
ubuntu@ip-xxxxxxx:~/zcash/src$ ./zcash-cli getinfo
error: couldn’t connect to server
ubuntu@ip-xxxxxxxxx:~/zcash/src$ ./zcash-cli getinfo
error: couldn’t connect to server
ubuntu@ip-xxxxxxxx:~/zcash/src$ ./zcash-cli getinfo
error: couldn’t connect to server
I will post my testing result based on Beta1.
Same waiting for it to be released 
It’s genproclimit, not procgenlimit.
1 Like
Test on C4.2xlarge:
ubuntu@ip-xxxxxxxxxx:~/zcash/src$ time ./zcash-cli zcbenchmark solveequihash 10
[
{
“runningtime” : 56.25224500
},
{
“runningtime” : 28.66505300
},
{
“runningtime” : 41.76247000
},
{
“runningtime” : 28.50802500
},
{
“runningtime” : 28.60153800
},
{
“runningtime” : 55.77457600
},
{
“runningtime” : 55.71295300
},
{
“runningtime” : 56.03348500
},
{
“runningtime” : 56.30824900
},
{
“runningtime” : 42.00046400
}
]
real 7m29.623s
user 0m0.004s
sys 0m0.000s
Another test with genproclimit parameter:
ubuntu@ip-xxxxxxxxx:~/zcash/src$ time ./zcash-cli zcbenchmark solveequihash 10 [
{
“runningtime” : 29.06952500
},
{
“runningtime” : 35.92974000
},
{
“runningtime” : 43.18478200
},
{
“runningtime” : 57.58049600
},
{
“runningtime” : 57.09156600
},
{
“runningtime” : 14.49394600
},
{
“runningtime” : 57.23365000
},
{
“runningtime” : 14.68988600
},
{
“runningtime” : 43.24290800
},
{
“runningtime” : 57.61352100
}
]
real 6m50.134s
user 0m0.004s
sys 0m0.000s
Regarding the results, I have a question, where are there ten “runningtime”? My cloud machine only has 4 vCPU.
The ‘10’ parameter you entered is for the number of consecutive times the equihash benchmark executes. There’s an optional second parameter for specifying the number of CPUs / cores but I don’t know if it’s actually implemented yet…
Do you mean the parameter “genproclimit”, I have set up this parameters, it works well.
Hi, @Shawn, I finally got round to doing this. I thought it would be a pain, but it was laughably simple 
! All I had to do was change it from
class="wikitable"
to
class="wikitable sortable"
and it automagically did what I wanted! Now you can sort the tables by column, which makes it much easier to see what performs best and worst. The method of entering data etc. (in fact, nothing else) has changed!
1 Like
@foodforthought Nice find! I didn’t even know about the different classes of tables, this will help a lot 
1 Like
Have made some modifications to the table; hope that’s cool with you? Have added RAM per core, and made the numerical numbers sortable as numbers, rather than strings.
I’m not sure everyone’s following the same formula for calculating hashrate/core, though !
1 Like
Yeah, I’m not sure we can use the H/s formula for a huge Cloud instance. Maybe we need to have a separate chart for clouds since you are using so many separate processors it’s probably not possible to accurately measure single core performance, nor is it possible to know the precise memory configuration. So it would only need the average time column, num. CPU’s, memory amount, and maybe a cost?