joncampbell123 / dosbox-x

DOSBox-X fork of the DOSBox project
GNU General Public License v2.0
2.67k stars 379 forks source link

Good prospects for DosBOX-X to be included in a Linux bootable distro (from a similar case scenario) #2223

Open dodleh opened 3 years ago

dodleh commented 3 years ago

In my search for the most accurate x86 emulation, in many years (as far back as the early 2000's), I have seen many emulation solutions centered around different concepts (Bochs, VirtualPC, QEMM, VMWare, VirtualBox, DosBox). I still think that for older systems DosBox brings a very good mix of performance and features and DosBox-X is very, very good (even if I am picky with some aspects).

I have been recently going down the avenue of Linux-DosBox integration with some interesting results. Of course, I am not really new to Linux either, but the point is that a bootable Linux-DosBox distro integration seems very promising. I have tested DosboxDistro. While it is severly lacking in some ways, the more important being that it uses an obsolete Dosbox from around 2009, it is highly interesting, and I hope that DosBox-X also gets integrated in way, in the future, in that distro or another.

The good things:

The bad things:

Wengier commented 3 years ago

Thanks for the suggestions! I certainly agree that are advantages for doing so. It seems that DosboxDistro is still in BETA status and has not been updated for a while. On the other hand, it appears that from the description it tries to do things such as calculator, calendar, image viewing, music playback, internet browsing, etc, which are well within the scope of DOSBox-X but outside the scope of vanilla DOSBox, so it will certainly work better if DOSBox-X is used.

dodleh commented 3 years ago

It is true that DosboxDistro is in Beta and, just as well, it seems to be less focused on being configuration-friendly but more of an experience or showcase. However, there is a clear advantage in that it completely frees resources that make the emulation much better knit altogether. Even if, in a best case scenario, it is still far from a Hyper-V implementation, the idea of building something akin to a hypervisor that runs the emulation is the closest to a bare-metal emulation and there are many qualitative and quantitative advantages. There are still drawbacks and I am sure of them, but since emulation is mostly CPU bottle-necked and not overly graphics intensive (at least at this stage), i see a lot of promise on such an approach. Hopefully there will be more interest from others as well. It is far from easy, though, and I congratulate anyone willing to fight such an uphill battle. I mostly can help only with testing, other things are beyond my expertise.