Parallels on a 2020 13.3” MacBook Pro

[iRun26.2]

I have a Windows FPGA simulator that I want to run on a MacBook Pro.

Will Parallels work on the 4-core 2020 13.3” machine?

(Obviously the 8-core 2019 16” MacBook Pro would be ideal but I really want portability when I’m doing other tasks).

 

[JohnDoe12]

The 32 GB machine is supposed to be able to run multiple VMs. I'm sure you'll be fine running one VM.

I used to run Parallels (Windows 10) on my 2010 MBP which had 4 or 8 GB of RAM and an Intel Duo processor. Just imagine. Obviously not top class performance but it worked.

I think you should be fine

 

[casperes1996]

Of course it will run. You could run it on a MacBook (the 12” one)... Hell, you could run it on a 2008 MacBook Air. It wouldn’t run well, but the code would execute.

 

[Dhock_Holiday]

As someone that as actively testing Windows 10 on my MacBook Pro 16 inch I can say it heavily depends on your workload. Opening up Word documents or excel spreadsheets is fine but anything even remotely task heavy will bog the machine down. Firing up Chrome and a few tabs on my Win10 instance will get the CPU right up to 80C easily. Debian is a bit better but I still see the MBP running too hot for my liking but YMMV

 

[maflynn]

I have a Windows FPGA simulator that I want to run on a MacBook Pro.

What's the technical specifications for the software? It may "run" but it may not function in a way that you are satisfied, i.e., too slow. boot camp may be your best bet if its particularly intensive. I'm not familiar with FPGA simulator, so I can't say one way or another.

 

[donawalt]

Contrary to others I run Parallels on the 16" MacBook Pro and it runs great, even heavier loads. I notice it uses the AMD 5500M graphics card, so I suspect others may see issues if their Mac has a low-end graphics solution; but that's just a guess - since Energy/Activity Monitor shows it always using the Radeon card. Mine doesn't even turn fans on unless I am doing something SUPER heavy, and even then only for a short time. I also noticed that it appears a lot more responsive after I tweaked some settings. The biggest was most apps, especially edge, had a pause/hesitation to it when it opened, I adjusted Settings > Network & Internet > Proxy, toggle off "Automatically detect settings and Use a proxy server." That made a huge difference.

 

[matram]

I had difficulty running Visual Studio under Win 10 on my maxed out 2016 15”. The 16” with max CPU / GPU and 32 GB RAM is the first MBP to give me acceptable performance.

As @Dhock_Holiday says it depends a lot on your workload.

 

[crispy86]

I had difficulty running Visual Studio under Win 10 on my maxed out 2016 15”. The 16” with max CPU / GPU and 32 GB RAM is the first MBP to give me acceptable performance.

As @Dhock_Holiday says it depends a lot on your workload.

Hi, I am looking to buy my first MacBook Pro. I would also like to run Visual Studio 2019 on Windows via Parallels . I was looking at the base Model MacBook Pro with 16gb. Do you think this will Be enough for it to run efficiently ?

 

[matram]

I use four virtual cores for the VM, the base six core may work if you only need one VM. VS is a 32-bit app. So 8 GB for the VM is adequate. I use a second VM for testing, that pushed me to eight cores and 32 GB to handle two VM and still have resources left over for MacOS.

 

[KPOM]

I have a Windows FPGA simulator that I want to run on a MacBook Pro.

Will Parallels work on the 4-core 2020 13.3” machine?

(Obviously the 8-core 2019 16” MacBook Pro would be ideal but I really want portability when I’m doing other tasks).

I’m not into gaming, but I do run Windows 10 in a virtual machine on my 2020 13” Ice Lake MacBook Pro. It works very well for what I need it (Quicken for Windows, Microsoft Edge for sites that don’t like Safari).

 

[crispy86]

I use four virtual cores for the VM, the base six core may work if you only need one VM. VS is a 32-bit app. So 8 GB for the VM is adequate. I use a second VM for testing, that pushed me to eight cores and 32 GB to handle two VM and still have resources left over for MacOS.

Thanks. I was also wondering. I will mostly be using Visual Studio for development work in my own time and for that reason I will be building .net core apps. Have you used Visual Studio on Mac at all?

 

[iRun26.2]

Of course it will run. You could run it on a MacBook (the 12” one)... Hell, you could run it on a 2008 MacBook Air. It wouldn’t run well, but the code would execute.

I actually have run it on a 12” MacBook. It runs incredibly slow.

I’m hoping for better results with the 2020 13.2” MBP with 4-cores relative to 2-cores (and 32G vs 8G RAM). My software looks like it is maxing out two cores when running on a 4-core machine.
[automerge]1590372633[/automerge]

What's the technical specifications for the software? It may "run" but it may not function in a way that you are satisfied, i.e., too slow. boot camp may be your best bet if its particularly intensive. I'm not familiar with FPGA simulator, so I can't say one way or another.

I’m hoping to avoid boot camp because I’d like to use other OS X apps while the simulator is running. Then I’d probably have to install MATLAB in both Windows (boot camp) and OS X. This would be wasteful.

 

[BigMcGuire]

I used Parallels on a 2017 MBP 13’ with 16GB of ram and the mid-range CPU option at the time. Visual Studio 2019 compiling 200k+ files worked very nicely. Couldn’t play games comfortably (even with an EGPU) but ... it worked well. I did this for a year+.

 

[matram]

Visual Studio 2019 compiling 200k+ files worked very nicely.

Maybe I should clarify that my issues with VS was not compile time but UI lag.

My application is about 70k lines and 200+ files. Compile time was OK.

I am using Jetbrains Resharper writing in C#, the "intellisense" (mostly from Resharper as I have disabled a lot of VS own) frequently looked up the UI on my old 2016, so my typing was 5+ characters ahead of what was shown in the editor, I often had to wait for the editor to catch up so I could see what I had written, very irritating, especially as I am not very fast typist.

The 8-core 16" seemed to be much better in handling these short burst of high CPU use, so the UI is now fluid, very happy with the 16". I do think Jetbrains realize that Resharper has performance issues and are working on those.

On a related note, before the 2016 15", I had a 2014 13". That was for me a mistake. The screen was too small for VS and I could clearly feel the lack of CPU power.

 

[UBS28]

edit: nvm

 

[iRun26.2]

What's the technical specifications for the software? It may "run" but it may not function in a way that you are satisfied, i.e., too slow. boot camp may be your best bet if its particularly intensive. I'm not familiar with FPGA simulator, so I can't say one way or another.

I’m not certain what the ‘technical specification’ for my FPGA simulator is. When running it on a 4-core Windows machine it is heavily using two of the eight threads (near 100%). Once in a while there is a spike in all eight threads. I’m sorry if that isn’t helpful.

 

[maflynn]

I’m not certain what the ‘technical specification’ for my FPGA simulator

The software you own should have what type of hardware is the minimum specifications, including CPU, ram, and GPU. Without knowing that information its hard to say what Mac will best fit and/or if parallels would be an ideal solution as opposed to a dedicated windows machine, or just using bootcamp.

 

[matram]

I’m hoping for better results with the 2020 13.2” MBP with 4-cores relative to 2-cores (and 32G vs 8G RAM). My software looks like it is maxing out two cores when running on a 4-core machine.

I’m not certain what the ‘technical specification’ for my FPGA simulator is. When running it on a 4-core Windows machine it is heavily using two of the eight threads (near 100%). Once in a while there is a spike in all eight threads. I’m sorry if that isn’t helpful.

This indicates that your FPGA simulator may not parallelize well. It may be written to only effectively use two threads.

If this is the case more cores will make (almost) no difference, it may even be negative as more cores tend to mean lower clock rates. For this type of application you should look for highest clock rate / best single thread performance and good thermals so you can use turbo boost as long as possible, for me this sounds more like a base 16".

Once a long time ago, I was responsible for procuring computer system for all type of simulations for a car OEM. In this situation we tried to understand what kind of computer architecture would give the best performance for a certain type of analysis. For instance FEM analysis required few very fast CPU because it parallelized very poorly. On the other hand CFD parallelized very well, so we could use simpler CPU but many many more of them.

Maybe you could research your application a bit to understand how it works and what would benefit you, in that way you may get better advice.

You may also want to google "Amdahls law" to understand a bit about how parallelization works, or sometimes does not work.

 

[iRun26.2]

This indicates that your FPGA simulator may not parallelize well. It may be written to only effectively use two threads.

If this is the case more cores will make (almost) no difference, it may even be negative as more cores tend to mean lower clock rates. For this type of application you should look for highest clock rate / best single thread performance and good thermals so you can use turbo boost as long as possible, for me this sounds more like a base 16".

Once a long time ago, I was responsible for procuring computer system for all type of simulations for a car OEM. In this situation we tried to understand what kind of computer architecture would give the best performance for a certain type of analysis. For instance FEM analysis required few very fast CPU because it parallelized very poorly. On the other hand CFD parallelized very well, so we could use simpler CPU but many many more of them.

Maybe you could research your application a bit to understand how it works and what would benefit you, in that way you may get better advice.

You may also want to google "Amdahls law" to understand a bit about how parallelization works, or sometimes does not work.

Could one even imagine that the 2020 13.3" being better because its of 30% single core performance benchmark over the 2019 16"? I suppose the better thermals of the 16" would override any performance gained in clock speed. The 30% better single core speed improvements probably are also quoted as going against the i9 (8-core) version of the 16" instead of the 6-core version you quoted.

The FPGA simulator software I use is pretty old so it makes sense that it is not optimized for parallel processing (it came for free with my other development tools). I've been unable to find a system requirement quote on the exact software I use but the student version quotes: Operating System: Windows XP/Vista/7/8. Memory (RAM): 512MB of RAM required. Hard Disk Space: 500MB free space required.

 

[casperes1996]

Could one even imagine that the 2020 13.3" being better because its of 30% single core performance benchmark over the 2019 16"? I suppose the better thermals of the 16" would override any performance gained in clock speed. The 30% better single core speed improvements probably are also quoted as going against the i9 (8-core) version of the 16" instead of the 6-core version you quoted.

Depends how long running the tasks you put it too are. If they are millisecond bursts, then yes, but anything that takes a bit to finish, no, since the 16" would turbo longer - it also has more cache, which could greatly benefit some tasks,

The FPGA simulator software I use is pretty old so it makes sense that it is not optimized for parallel processing (it came for free with my other development tools). I've been unable to find a system requirement quote on the exact software I use but the student version quotes: Operating System: Windows XP/Vista/7/8. Memory (RAM): 512MB of RAM required. Hard Disk Space: 500MB free space required.

That really doesn't sound like it's that intensive. - What program is it? Unless you need to use it specifically because of workflow or integration with other tools you may be able to find an alternative that does the same thing but uses the system better to finish faster.