Is this an emulation layer? How is the efficiency? And could it be used directly on Windows 10? I believe They will lose most of their computing power if no x86 is available

---

No. I use WSL1
https://en.wikipedia.org/wiki/Windows_Subsystem_for_Linux

Thank you but how does that solve the ARM architecture-only target Hoarfrost I assume is saying? Need emulation layer. If no target running simultaneously that is x86 compatible, their compute power will go way down.

---

I have noticed that cmdock has a job file for x86-64 pc cmdock-l at https://www.sidock.si/sidock/download/cmdock-l_job_2.02_x86_64-pc-linux-gnu.xml and https://www.sidock.si/sidock/download/cmdock-l_job_2.02_windows_x86_64.xml
When i have compared them with https://www.sidock.si/sidock/download/cmdock-s_job_2.02_arm-unknown-linux-gnueabihf.xml cmdock they were almost identical.
They even use same command line parameters. <command_line>-c -j 1 -b 1 -r target.prm -p "$PWD\data\scripts\dock.prm" -f htvs.ptc -i ligands.sdf -o docking_out</command_line>

Thank you but how does that solve the ARM architecture-only target Hoarfrost I assume is saying? Need emulation layer. If no target running simultaneously that is x86 compatible, their compute power will go way down.

Yes you are right Sidock will lose computing power if no work is available. I am reasonably confident that the power will return when work is available

If going to ARM-only architecture is true, seems like a chicken before the egg scenario.. they have to attract people with ARM devices before switching over otherwise echoes until people realize. I hope Hoarfrost post was out of context somehow and it's not happening. But if it is please blog post the reasoning why.

---

If going to ARM-only architecture is true, seems like a chicken before the egg scenario.. they have to attract people with ARM devices before switching over otherwise echoes until people realize. I hope Hoarfrost post was out of context somehow and it's not happening. But if it is please blog post the reasoning why.

It's the same source code and data format. You just need to compile from source for your native architecture (x86 in this case). If you compile for your native OS too, the resulting binary executes the tasks natively and there is no emulation involved. In theory, there is nothing preventing the project from releasing official x86 binaries for the short tasks on both Linux and Windows, just like other projects that supports multiple architectures. kotenok2000 uses WSL1 on Windows likely because the host is Windows but the compiled binary is for Linux. I followed his guide for building the binaries on Linux but my host is Linux, so I've been running the short tasks natively. I did this mostly because the previous long tasks were too long for my preference. So before Ebola, my x86 hosts exclusively ran short tasks. The runtime of Ebola is totally fine though and hopefully the next long task won't be day long again.

If you manage to get it compile under Windows, it will be a native Windows application, removing any need of Linux VM, including WSL.

I don't think the post was out of context. To me it says exactly what it reads. There will be a period of time where there are only going to be tasks available ARM devices

An explanation is needed, of course. At this moment we have two applications: "Longs tasks" and "short tasks". Applications identical by sense and source code. For "Long tasks" we perform generation of tasks for main search that need much more CPU time per task. We send these to computers with x86-64 platform. Workunits for "Short tasks" much more shorter due to properties of target and search goals. We send these to ARM-based computers.
Now don't have another target right now.
After ending of "ebola_GP_v1" we will try return to generation of "Short tasks" for x86-64 platform, but presumably, they won't be enough for full load.

Will you have another vote for the next target?

Thank you, Hoarfrost It sounds like it would not be worth the hassle recompiling for x86 as there will not be much workload in any case. You have anticipation of when lots of x86 work will resume?

---

Fortunately there are precompiled binaries.
https://gitlab.com/Jukic/cmdock/-/releases

I can throw my Raspberry Pi's on the project for a while to help get through those ARM tasks.

You have anticipation of when lots of x86 work will resume?

I don't know. We doesn't have next target, simply.

So, when will the next job be available? 🤷‍♂️

---

non-profit org. Play4Life in Zagreb, Croatia, EU

I'm not sure when but it will be available when it starts getting distributed to people's computers :-)

I propose as one of the voting options what affects aging. For example, Advanced glycation end-products (AGEs), more details about the candidates here

https://bostonmatrix.org

https://www.mayoclinic.org/drugs-supplements/collagenase-clostridium-histolyticum-aaes-subcutaneous-route/description/drg-20490865