Re: Benchmarking DOSBox on latest machines

Posted on 2019-10-19, 05:52

by

by jmarsh

in PC Emulation

in PC Emulation

I'd be interested to try a 64-bit Visual Studio build, but my attempts at building one have resulted in executables that do not work. If you're using the config.h found in src/platform/visualc, it needs some fixes for 64-bit. Remove the "#define C_TARGETCPU X86" line, set C_FPU_X86 to 0 and change …

---

Re: 64-bit dynamic_x86 (patch)

Posted on 2019-10-16, 09:37

by

by jmarsh

in DOSBox Development

in DOSBox Development

Maybe it's time to ruin everyone's benchmarking results by improving the 32-bit core...

---

Re: Internal USB header

Posted on 2019-10-16, 04:39

by

by jmarsh

in Marvin \ General Old Hardware

in Marvin \ General Old Hardware

If you're sure it uses the same pin configuration as your header (there are two common different layouts...) you can poke a hole through the blocked plug.

---

Re: 64-bit dynamic_x86 (patch)

Posted on 2019-10-15, 14:00

by

by jmarsh

in DOSBox Development

in DOSBox Development

The dynamic core always uses FSAVE/FRSTOR regardless of the host CPU's capabilities, plus the Dhrystone benchmark is specifically designed to not test floating point performance.

---

Re: We Need to Preserve the Vanilla Quake Experience for Modern Systems

Posted on 2019-10-14, 20:09

by

by jmarsh

in DOS

in DOS

It supports 320x240, 400x300, 640x480 and a bunch of other modes in between and beyond.

---

Re: We Need to Preserve the Vanilla Quake Experience for Modern Systems

Posted on 2019-10-14, 10:57

by

by jmarsh

in DOS

in DOS

If you want to play with a stretched hud, power to you. But the artistically intended option shouldn't be so limited in its availability. Original DOS Quake let you choose a bunch of resolutions besides 320x200 and practically all of them resulted in a stretched HUD, menus, console etc. The …

---

Re: 64-bit dynamic_x86 (patch)

Posted on 2019-10-14, 10:43

by

by jmarsh

in DOSBox Development

in DOSBox Development

That's why I said where the majority of execution time is spent during that benchmark.

---

Re: 64-bit dynamic_x86 (patch)

Posted on 2019-10-14, 07:39

by

by jmarsh

in DOSBox Development

in DOSBox Development

Dynamic code (where the majority of execution time is spent during that benchmark) is assembled at run-time so the compiler doesn't matter. Those results look more like normal/dynrec core vs. dyn_x86 - which is very possible since 0.74-3 uses a different .conf file than an SVN build.

---

Re: 64-bit dynamic_x86 (patch)

Posted on 2019-10-14, 04:08

by

by jmarsh

in DOSBox Development

in DOSBox Development

robertmo wrote: it looks 32-bit windows build is also affected with Ryzen cpus. Benchmarking emulators on latest machines though "a fair amount […] Show full quote it looks 32-bit windows build is also affected with Ryzen cpus. Benchmarking emulators on latest machines though "a fair amount of …

---

Re: 64-bit dynamic_x86 (patch)

Posted on 2019-10-12, 19:08

by

by jmarsh

in DOSBox Development

in DOSBox Development

Trying to emulate an entire 32-bit x86 CPU on a 32-bit x86 CPU is awkward because most of it is unusable by user code; segment registers, system registers, even ESP is restricted from being freely modifiable. A 64-bit CPU has an extra 8 general purpose registers to help get the job done.

---

Re: ethflop - a network-backed floppy emulator for DOS

Posted on 2019-10-09, 13:01

by

by jmarsh

in Release Announcements

in Release Announcements

TFTP is so basic it doesn't allow random reads/writes, only whole files can be transferred.

---

Re: DOSBox Feature Request Thread

Posted on 2019-10-08, 13:44

by

by jmarsh

in DOSBox General

in DOSBox General

Most DOSBox recordings need to be edited afterwards and there's not a lot of support for editing Matroska, plus the only way to store ZMBV data in it would be to use VFW codec mode which gives no benefits over AVI.

---

Re: 386 system came with "lost" audio authoring software?

Posted on 2019-10-08, 06:23

by

by jmarsh

in Marvin \ Sound

in Marvin \ Sound

Template (by Judy S. Kravitz) It is a program that apparently is used to output a midi file to a dot matrix printer that also punches holes in the paper for some type of player instrument.. maybe an organ? A player piano. The roll is held on two spools surrounding a bunch of intake valves, and you …

---

Re: hard drive type, sector ect settings, and sector size

Posted on 2019-10-08, 06:14

by

by jmarsh

in Marvin \ General Old Hardware

in Marvin \ General Old Hardware

I've never seen a bios with a sector size setting. Anything new enough to cope with 4K sectors would be able to auto-detect. All old hard drives use 512 byte sectors, and so do all SD cards.

---

Re: Windows 10: Can't get fullscreen to work

Posted on 2019-10-08, 03:52

by

by jmarsh

in DOSBox General

in DOSBox General

change output to opengl in the .conf file. Also set aspect to true if you don't want the original 4:3 screen stretched on a widescreen monitor.

---

Re: Best way to shutdown a 386/486 ?

Posted on 2019-10-07, 14:48

by

by jmarsh

in Marvin \ General Old Hardware

in Marvin \ General Old Hardware

Surely the heads float in the air when the unit is powered off? It's only when the drive is moved that you run the risk of them crashing into the platters. The heads are spring loaded to be pushed towards the platters, they only float due to air passing between them and the surface. When it stops …

---

Re: PowerPC Dynamic Recompiler (patch)

Posted on 2019-10-06, 17:03

by

by jmarsh

in DOSBox Development

in DOSBox Development

Patch in first post updated, should hopefully work with OSX now. Edit: And again, because apple's gcc confuses comments and instructions.

---

Re: "Non-system Disk Or Disk Error" after overwriting files on an MS-DOS partition

Posted on 2019-10-06, 11:00

by

by jmarsh

in Marvin \ General Old Hardware

in Marvin \ General Old Hardware

Either IO.SYS or MSDOS.SYS (or one then the other, I forget the details) have to be located first in the root directory entry. Blindly copying files doesn't ensure they will end up in the same order.

---

Re: 64-bit dynamic_x86 (patch)

Posted on 2019-10-06, 07:36

by

by jmarsh

in DOSBox Development

in DOSBox Development

You can look at the code for yourself; the only difference in risc_x64.h for windows vs. linux is that windows requires extra instructions to be emitted to save/restore RSI and RDI all the time. Executing more instructions by definition can't increase performance, the difference is likely due to …

---

Re: 64-bit dynamic_x86 (patch)

Posted on 2019-10-06, 05:47

by

by jmarsh

in DOSBox Development

in DOSBox Development

64-bit Linux in general should be faster than windows due to a more sensible ABI.

---