VOGONS

Common searches


Fastest CPU for DOSBox

Topic actions

First post, by senrew

User metadata
Rank Oldbie
Rank
Oldbie

Tried searching and paging through posts, but I couldn't find anything that addressed this directly.

What would be the absolute fastest single core CPU that would benefit DOSBox? From what I can tell, DOSBox itself can't use multiple cores even if auxilliary processes can (SDL, DirectSound, etc)

Any suggestions?

Halcyon: PC Chips M525, P100, 64MB, Millenium 1, Voodoo1, AWE64, DVD, Win95B

Reply 1 of 20, by Dominus

User metadata
Rank DOSBox Moderator
Rank
DOSBox Moderator

The fastest you can get 😉

Windows 3.1x guide for DOSBox
60 seconds guide to DOSBox
DOSBox SVN snapshot for macOS (10.4-11.x ppc/intel 32/64bit) notarized for gatekeeper

Reply 2 of 20, by sliderider

User metadata
Rank l33t++
Rank
l33t++
senrew wrote:

Tried searching and paging through posts, but I couldn't find anything that addressed this directly.

What would be the absolute fastest single core CPU that would benefit DOSBox? From what I can tell, DOSBox itself can't use multiple cores even if auxilliary processes can (SDL, DirectSound, etc)

Any suggestions?

The AMD Sempron 150 runs circles around the fastest single core Athlon64 FX and Pentium4 EE chips in Passmark so as far as single core CPU's that you can still buy new are concerned, that would probably be near the top of the list.

Reply 3 of 20, by senrew

User metadata
Rank Oldbie
Rank
Oldbie

Ok, let's take this the other direction. Will a modern multi-core machine be better suited to running DOSBox? The extra cores wouldn't be utilized, but with something like TurboBoost, combined with the fact that a modern CPU would be more efficient in general, shouldn't it still be better to just say fuck it and run DOSBox on whatever fastest modern machine I have? So...Modern beast CPU vs Specialized single-core?

Halcyon: PC Chips M525, P100, 64MB, Millenium 1, Voodoo1, AWE64, DVD, Win95B

Reply 4 of 20, by Dominus

User metadata
Rank DOSBox Moderator
Rank
DOSBox Moderator

Yes

Windows 3.1x guide for DOSBox
60 seconds guide to DOSBox
DOSBox SVN snapshot for macOS (10.4-11.x ppc/intel 32/64bit) notarized for gatekeeper

Reply 5 of 20, by senrew

User metadata
Rank Oldbie
Rank
Oldbie

Ok, good enough for me. Lock this thread please. Maybe if someone else has this question the brevity of this thread will help them.

Halcyon: PC Chips M525, P100, 64MB, Millenium 1, Voodoo1, AWE64, DVD, Win95B

Reply 6 of 20, by gulikoza

User metadata
Rank Oldbie
Rank
Oldbie

There's no reason DOSBox wouldn't benefit from extra cores...there's a bunch of stuff that runs in parallel. It's true that 1 thread may do most of the processing, but DOSBox 0.74 spawns 18 (!) threads in total on my machine. Alongside with that you still have the OS running, AV software...

http://www.si-gamer.net/gulikoza

Reply 8 of 20, by robertmo

User metadata
Rank l33t++
Rank
l33t++
gulikoza wrote:

DOSBox 0.74 spawns 18 (!) threads in total on my machine.

what are they? you need to get 10 core xeon 😀 i guess it can even do 20 threads so that covers even OS running and AV software 😉

BTW i wonder how much have you got here Suggested specs for a full speed DOSbox machine?

Reply 9 of 20, by gulikoza

User metadata
Rank Oldbie
Rank
Oldbie

It doesn't mean all 18 threads do something useful 😀. Most of them are created by the OS and/or SDL (even a video driver like OpenGL can use several threads internally) and used for who knows what. But MIXER_CallBack() runs in it's own thread (that's how SDL is designed) and various patches add a small amount of additional threading (D3D, mt32...even when running Munt as a midi device, it will use a different core). That means that side-by-side (same mhz/arch) a multicore cpu will always be a few % faster no matter what (even more so when all the OS crap can be offloaded to a different core).

So as Dominus said, searching for the fastest single core cpu because dosbox can only use one is quite pointless. Times when single core CPUs were faster than multicore are (fortunately) long behind us 😀

http://www.si-gamer.net/gulikoza

Reply 10 of 20, by alvaro84

User metadata
Rank Member
Rank
Member

I have a somewhat different question: has anyone studied how new architectures work with DosBox? I remember that in the old days P4s were miserably weak under DosBox compared to the AMD lineup. Later the Core CPUs became good at DosBox, and lately it seems that AMD Bulldozer and its successors are a huge step back on AMD's side. I've seen tests where Bulldozers did less then half clock-to-clock compared to Sandy/Ivy Bridge processors. Now, I'm curious if Trinity and the future Kaveri can run DosBox better, because as things are now I'm really afraid that these new APUs couldn't even catch my old Core 2 duo 🙁 Which would be a shame, because otherwise they'd provide nice small form factor rigs with quite decent graphics.

Shame on us, doomed from the start
May God have mercy on our dirty little hearts

Reply 11 of 20, by Sol_HSA

User metadata
Rank Member
Rank
Member
alvaro84 wrote:

I have a somewhat different question: has anyone studied how new architectures work with DosBox? I remember that in the old days P4s were miserably weak under DosBox compared to the AMD lineup.

The reason for that was P4's 100-mile long pipeline, which suited (most) native x86 binaries just fine.

Fortunately for dosbox, there's plenty of VM's in use today - javascript engines, flash, java, .net, etc. and having a good performance on those will also have a good performance in dosbox.

http://iki.fi/sol - my schtuphh

Reply 12 of 20, by alvaro84

User metadata
Rank Member
Rank
Member

Yet it seems either VMs weren't in focus when they made the Bulldozer design, or there's something in DosBox in particular, but this doesn't only make BD ugly, it's a complete disaster:

http://pctuning.tyden.cz/hardware/procesory-p … sktopu?start=13

Haswell runs DosBox 2.5 times(!) faster than BD (clock-to-clock). This is why not having a Trinity test bothers me, but I don't expect a huge improvement. I also couldn't find a comparable test with older CPUs, so I'm somewhat in the dark.

I can only hope they keep to the habit of testing with DosBox at least until Kaveri arrives. That's said to have changes in its architecture to alleviate a lot of bottlenecks (like they'll double its decoders). We'll see if the problem is there, or somewhere else. BD (Trinity, Kaveri) is rumored to have quite a long pipeline anyway (I haven't seen any official data about it). Probably not as long as the infamous Prescott, but surely longer than K10 or Core designs.

And that long pipeline didn't work too well, it had a very unbalanced performance. It could have been decently strong (and outstanding in a few areas) if they could double its clock rate (they planned a 10GHz version later), but it proved impossible.

Bulldozer seems to be in the same trap: it's unbalanced and looks to be designed for higher clocks which it can't reach. My question is that if it's because of a long pipeline, or other bottlenecks? If it's the former, they'll have to do something to rev it up within a reasonable thermal envelope (energy consumption), or look for some other design (Jaguar, maybe?). If the latter, they can still hone the existing design to make it better.

(Because it looks very interesting. That shared FPU and the future common address space with the IGP never ceases to amaze me!)

Shame on us, doomed from the start
May God have mercy on our dirty little hearts

Reply 14 of 20, by Mau1wurf1977

User metadata
Rank l33t++
Rank
l33t++

That is funny 😀 Amazing how much faster Intel is.

My website with reviews, demos, drivers, tutorials and more...
My YouTube channel

Reply 15 of 20, by alvaro84

User metadata
Rank Member
Rank
Member

I agree. A newer build would be good. Of course the most interesting part is the dynamic core. What I know, my old Core2 Duo (Conroe 4M, overclocked @ 3.1GHz, with DDR2-800 and a Radeon 6670) can do like 92fps with quake timedemo1 in 320x200, run in a 2013/04 Ykhwong build. But I don't know anything about these Czech guys' tests, not even the resolution 🙁

If it was apples to apples, the result would look grim dark for the new AMD parts (hell, my rig is over 6 years old!). But it's a very different build to begin with.

Shame on us, doomed from the start
May God have mercy on our dirty little hearts

Reply 17 of 20, by Mau1wurf1977

User metadata
Rank l33t++
Rank
l33t++

Interesting...

So I quickly ran Quake shareware timedemo demo1 on three of my machines. Standard DOSBox, Sound Blaster disables, 800 x 600 window, opengl, aspect, dynamic and max cycles.

AMD Phenom II 555: 96 fps
Intel Pentium 2.9 GHz: 101 fps
Core i7: 157.4 fps (59.7 fps in SVGA)

My website with reviews, demos, drivers, tutorials and more...
My YouTube channel

Reply 19 of 20, by Mau1wurf1977

User metadata
Rank l33t++
Rank
l33t++

Celeron 1.7 notebook (I think it's Sandy Bridge): 62.2 fps

The i7 is quite new, 3770.

My website with reviews, demos, drivers, tutorials and more...
My YouTube channel