it absolutely doesn't. you should be using the viewer only. you have no need for their server. you have Vino running on your system which is the server.

I use it every day with my systems and have no such account.

if this is somehow something new in the latest version, you can use a slightly older version.

i dropped mine on my google drive here: https://drive.google.com/open?id=1_GacMnwybJ-EqQAh_f-ff5njsh2zlxfx

Hi Ian,

I believe that I read that it is something new they started. I did get just the viewer. I'didn't even know it was RealVNC Viewer until after I ran it, or maybe I did and figured I would have no problem with it. I'll download the one you have and see what happens.

Have a great day! :)

Siran

---

CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath

While I have been playing with the "other" MB trying to see how many gpus I could get in to recognize (I am stuck at 8 on a 7 slot board :) I basically robbed my highest performing machine of its entire gtx 1060 3GB supporting cast.

So its running only three Gtx 1070Ti's. While my RAC graph is showing a slight decline it certainly hasn't fallen off the cliff when I deprived it of between 4 and 6 gtx 1060's.
Maybe a couple more days will make a difference?

Tom

---

A proud member of the OFA (Old Farts Association).

Greetings All

Just another idea for remoting into computers, I use a program called NoMachine, that I can remote into my PCs be it a Win or Linux Box.

Just an idea, easy to use, but then I havent done anything Raspberry PI setups.

Regards

Greetings,

I now have access to my new Linux box and my 3 Pis through VNC. Thank you Ian for the older version of the VNC viewer.

Now, I have a question concerning my switch. I'll be going from my 4 port (or 5 depending on how you look at it) to an 8 port (or 9 port depending...). My Pis and the Linux box get the same IP addresses and the IPs are consecutive from Pi 1, 2, 3 and Linux box. When I switch to the new switch, (switch to the new switch, get it ;) ), if I start the Pis and the box in the same order that I initially started them they will get the same IP address, correct?

I'm assuming that there is circuitry (database) in the switch that retains the IPs per device. I did not have my Pi 3 plugged in when I had the Linux box plugged in and the box got the next higher IP then the Pi 3 has. Ok, wait just a minute. The switch does not assign IP addresses, my Ethernet router does per MAC address. So it maintains the IPs for each device. I believe I just answered my own question. :)

Thanks anyway for any comments on my deductive reasoning or whatever. ;) lol

Once again, thank you Ian for the viewer.

Have a great day! :)

Siran

---

CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath

Greetings,

I now have access to my new Linux box and my 3 Pis through VNC. Thank you Ian for the older version of the VNC viewer.

Now, I have a question concerning my switch. I'll be going from my 4 port (or 5 depending on how you look at it) to an 8 port (or 9 port depending...). My Pis and the Linux box get the same IP addresses and the IPs are consecutive from Pi 1, 2, 3 and Linux box. When I switch to the new switch, (switch to the new switch, get it ;) ), if I start the Pis and the box in the same order that I initially started them they will get the same IP address, correct?

I'm assuming that there is circuitry (database) in the switch that retains the IPs per device. I did not have my Pi 3 plugged in when I had the Linux box plugged in and the box got the next higher IP then the Pi 3 has. Ok, wait just a minute. The switch does not assign IP addresses, my Ethernet router does per MAC address. So it maintains the IPs for each device. I believe I just answered my own question. :)

Thanks anyway for any comments on my deductive reasoning or whatever. ;) lol

Once again, thank you Ian for the viewer.

Have a great day! :)

Siran

I believe, that the addresses might change when the Ethernet router is rebooted, depending on the order it finds them assuning they are switched on. Certainly the address you get from your ISP will change on reboot, unless you pay for a fixed IP address, this is usually only available for business users.

I would just set static IPs on the devices so they never change. Or setup a DHCP reservation in your router for each device so they always get assigned the same IP.

If you do static IPs on the device, make sure you restrict your DHCP range and choose static IPs outside of that range.

For example on my home network, my router’s DHCP range is limited from 2-100. Anything with static IPs I give them IPs > 100.

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

I believe, that the addresses might change when the Ethernet router is rebooted, depending on the order it finds them assuning they are switched on. Certainly the address you get from your ISP will change on reboot, unless you pay for a fixed IP address, this is usually only available for business users.

He’s talking about his LAN IPs not his WAN IP.

WAN IP doesn’t always change from a router reboot either. Usually you have to disconnect your router for several minutes or connect a different device or go into the router settings and manually release/renew the lease.

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

My gateway hasn't changed its WAN address since I signed up with the ISP and it has been rebooted countless times and has had service out for an hour sometimes. The router has had the same gateway address too. Only do the DHCP clients occasionally change and only rarely through many router reboots. Never bothered with static IP addresses.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

My gateway hasn't changed its WAN address since I signed up with the ISP and it has been rebooted countless times and has had service out for an hour sometimes. The router has had the same gateway address too. Only do the DHCP clients occasionally change and only rarely through many router reboots. Never bothered with static IP addresses.

Consider yourself lucky. If for any reason I'm offline for around half an hour or so, they give me a new IP. Also at least once a year they force a new IP. I think they want to be sure if you need fixed IP you pay for fixed IP.

My gateway hasn't changed its WAN address since I signed up with the ISP and it has been rebooted countless times and has had service out for an hour sometimes. The router has had the same gateway address too. Only do the DHCP clients occasionally change and only rarely through many router reboots. Never bothered with static IP addresses.

Consider yourself lucky. If for any reason I'm offline for around half an hour or so, they give me a new IP. Also at least once a year they force a new IP. I think they want to be sure if you need fixed IP you pay for fixed IP.

Hi Gary,

We had the same IP for over 2 years before we upgraded the service and got a new one and have had it ever since just before Oct. 28, 2017.

Have a great day! :)

Siran

---

CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath

On the flip side, not having the ability to change it at will can be a pain.

Whenever the Chinese govt discovers my VPN they IP-ban me. So I need to change it to restore access. Luckily it’s not too difficult recently. And I’ve added additional layers of obfuscation that hopefully keep it hidden longer.

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

I've had the same WAN IP since 2007. Probably because I'm served out of CO that was built in the 60's. No fiber or fast Internet speeds here. Plain old aerial copper lines.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

My gateway hasn't changed its WAN address since I signed up with the ISP and it has been rebooted countless times and has had service out for an hour sometimes. The router has had the same gateway address too. Only do the DHCP clients occasionally change and only rarely through many router reboots. Never bothered with static IP addresses.

Consider yourself lucky. If for any reason I'm offline for around half an hour or so, they give me a new IP. Also at least once a year they force a new IP. I think they want to be sure if you need fixed IP you pay for fixed IP.

I'll second that, I cannot reconnect to my ISP if my internet router is disconnected or switched off for even a few seconds.

Usually the quickest way to restore the WAN internet connection is to switch off the router for at least a minute.

Greetings,

I do believe that the GT 740 SC GPU was slowing things down considerable. I have noticed, since I removed it to put into the Linux box which is no longer in it either, that my overall RAC and Windows PC exclusive RAC have increased past where they were when running the 740. Sooooooo... I will no longer be using it for SETI. I have packed it away in its pretty little box and shelved it for the time being in case I have a GPU blowout and need it just for video purposes only. :)

Have a great day! :)

Siran

---

CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath

Greetings,

I do believe that the GT 740 SC GPU was slowing things down considerable. I have noticed, since I removed it to put into the Linux box which is no longer in it either, that my overall RAC and Windows PC exclusive RAC have increased past where they were when running the 740. Sooooooo... I will no longer be using it for SETI. I have packed it away in its pretty little box and shelved it for the time being in case I have a GPU blowout and need it just for video purposes only. :)

Have a great day! :)

Siran

Question. You are running your cpu cores at 90%? I have seen my whole system slow down if I am running all cpu cores. It is possible if you where running all cores, running the gt 740 would slow the whole system down.

Discrete gpus do not "usually" slow down the rest of the system. I have no theory to account for your results unless it was the 100% thingy.

Tom

---

A proud member of the OFA (Old Farts Association).

Greetings,

I do believe that the GT 740 SC GPU was slowing things down considerable. I have noticed, since I removed it to put into the Linux box which is no longer in it either, that my overall RAC and Windows PC exclusive RAC have increased past where they were when running the 740. Sooooooo... I will no longer be using it for SETI. I have packed it away in its pretty little box and shelved it for the time being in case I have a GPU blowout and need it just for video purposes only. :)

Have a great day! :)

Siran

Question. You are running your cpu cores at 90%? I have seen my whole system slow down if I am running all cpu cores. It is possible if you where running all cores, running the gt 740 would slow the whole system down.

Discrete gpus do not "usually" slow down the rest of the system. I have no theory to account for your results unless it was the 100% thingy.

Tom

I agree with Tom. normally the problem is the CPU usage who make the entire host slower.

On my host when i run with hyper treading on (6 cores 12 threads) the sweet spot is 4 GPU + 6 CPU WU running at the same time anything above just makes the entire host production slower.

But be aware i not talking about RAC, i talking about the sum of the real crunching times x number of cores running x time. The RAC is not a good indicative of the short term host production. The RAC could take weeks to rely show the changes.

Of course YMMV, each host combination is unique. So test is needed.

if you wish to know this is how we made the test (Keith makes similar test on his host at the time)
(The example is with my host 4 GPU + CPU with up to 12 cores available)
- First you need to compare similar WU. SETI has very different type of WU Arecibo, Vlars, BLC etc.
If you not follow this the complete test is meaning less.
- If you use a program like project lasso to care about the affinity of the threads not forget to turn it on.
- Start you host with all the GPU's running and up to 50% of the CPU cores, remember each GPU WU uses a core.
- In my hosts that gives 4 GPU + 2 CPU
- Wait few WU where crunched on the CPU core, we use a mean of 10 WU but you could use a smaller sample.
- record the numbers, crunching time x cores used
- Increase the CPU core usage to 4 GPU's + 3 CPU
- repeat the process up to the host max 4 GPU +8 CPU

With the data now generate a table cores used x crunching time, you will notice there will be a sweet point where the production per hour reaches the maximum and above that it goes downs. That is your host sweet point. Above this point you only waste electric power and produces more heat. LOL

My test data shows in my host the increase of production from 4 to 6 CPU is minimal and the electric power & heat produced increaces a lot that's is why i ended running 4 GPU + 4 CPU WU.

I know is a tedious process but is the only one you could be sure what is the best point for your host.

Obviously anything else you run on the host at the time of test makes difference, so remember to look if any hidden process is running at the time. Windows could run several hidden process without warning you.

If you decide to do the test my suggestion is to do the test with and without the second GPU installed. As Tom posted is highly unlikely the total production of the host will be slower with a second discrete GPU. If your target is to increase the RAC the GPU production is normally bigger than the CPU, but that depends of what GPU x CPU you have of course, that's why i suggest both tests.

As always share your results if possible.

---

Greetings,

I do believe that the GT 740 SC GPU was slowing things down considerable. I have noticed, since I removed it to put into the Linux box which is no longer in it either, that my overall RAC and Windows PC exclusive RAC have increased past where they were when running the 740. Sooooooo... I will no longer be using it for SETI. I have packed it away in its pretty little box and shelved it for the time being in case I have a GPU blowout and need it just for video purposes only. :)

Have a great day! :)

Siran

Question. You are running your cpu cores at 90%? I have seen my whole system slow down if I am running all cpu cores. It is possible if you where running all cores, running the gt 740 would slow the whole system down.

Discrete gpus do not "usually" slow down the rest of the system. I have no theory to account for your results unless it was the 100% thingy.

Tom

Hi Tom,

I have <project_max_concurrent>8</project_max_concurrent> in my app_config.xml file. With the 740 I was running 6 CPU and 2 GPU WUs. Without I'm running 7 CPU and 1 GPU. Changing the "Use at most [xxx]% of the CPUs" changes nothing. It was set to 100%, I changed it to 75% and I'm still running 7 CPUs. I'm thinking that the 740 was having issues due to the settings in the command lines in app_config.xml.

My Astropulse GPU WUs are taking roughly 26-27 minutes on my 1050 Ti and took roughly 45 minutes on my 740 SC. I can put the 740 back in to verify this if needed if the one WU left on my list of Astropulse WUs goes away. It shows being assigned to device 1, the 740.

I don't know, all I know is that the system is better without the 740.

Have a great day! :)

Siran

---

CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath

I am very interested in faster cpu crunching on my https://setiathome.berkeley.edu/show_host_detail.php?hostid=8676008 box.

Currently it is averaging just below 2 hours per cpu task.

I read an assertion that while NUMA will speed up your frames / second a little bit (in gaming) but non-NUMA will speed up your "overall production" so I am trying it out.

Tom

---

A proud member of the OFA (Old Farts Association).

There is a lot about NUMA and non-NUMA over on:

Ryzen and Threadripper


Hope that helps!

Happy fast crunchin',
Martin

---

See new freedom: Mageia Linux
Take a look for yourself: Linux Format
The Future is what We all make IT (GPLv3)

There is a lot about NUMA and non-NUMA over on:

Ryzen and Threadripper


Hope that helps!

Happy fast crunchin',
Martin

Thank you. I have played with both AMD Numa/non-Numa and am now playing with Intel Numa/non-Numa.

The catch is the hardware and memory architecture are enough different that things that "oughta" work don't and vice versa.

Then you have to factor out the different processing times depending on your data mix.

So I am looking for the "average processing time" to drift lower and lower. Right now, it is dipping a little but probably due to the data mix, I was just getting.

Tom

---

A proud member of the OFA (Old Farts Association).