Discussion in 'Headphone Amplifiers and Combo (DAC/Amp) Units' started by purr1n, Oct 29, 2019.
I think Hel is worth a shot
Hel PSU doesn't fit into German (and other European) power outlets (there's only one PSU for worldwide distribution). Adapter is needed, not a big deal, though.
And Hel needs WASAPI push, WASAPI event causes distorted sound.
This on foobar20000? WASAPI event is famously, badly implemented on fb2k.
Oh sorry, I didn't know that. Event always worked for me with several DACs and Dirac, until Hel. I thought event should be preferred if it works, I got that info from the jriver documentation: https://wiki.jriver.com/index.php/WASAPI_Event_Style
Does this not apply in the same way for Foobar2000 (event is better presupposed it works)?
Like that wiki explains it can depend on hardware support but yes, if event works without problems it should be preferred over Push. I've been looking but can't find the thread about WASAPI event often being buggy on foobar
i work at a game studio, i could see something like this replacing all those Arctis headsets
I've built a simple amp around the OPA168EVM kit. Didn't sound bad at all. Using an RK27, it sounds pretty good really.
You can get the board here for just $25 direct from TI: http://www.ti.com/tool/OPA1688EVM?jktype=tools_software
I'm really glad to see Schiit using this line of hardware from TI. Really impressive stuff for chip amps.
Wanted to follow up with this comment. Jason calculated that the Bifrost 2 and higher priced DACs had 23ms in latency. Seemed high, as I just received a monitor with advertised 0.3ms response time (it's a newer generation 240hz monitor).
I cannot detect any noticeable sound delay after I take an action on my computer either via keyboard or mouse click. Played CSGO and some other games to test this. Not sure if there is a sound test/input latency test out there I could use. I have the Gungnir Multibit A1 hooked up.
So those worried about latency issues with a multibit DAC, I wouldn't sweat it. I have no problems thus far.
The D/S cards and Modi Mulitbit have 11ms latency according to Jason. I assume Modi 3, Fulla 3 and Hel should be the same.
FWIW in case there are those concerned with gaming use with a multibit DAC.
I actually have the answer to this issue.
It literally does not matter. The lowest average response time of a computer at 144hz and beyond is like...25ms. If you're lucky.
The cpu has to take in all the inputs, process them, work the game logic, then send that all to the gpu. The GPU goes through all the math allocated to it, as well as the image it needs to display.
So you end up the CPU time (7ms at 144fps) + GPU (another 7ms) + input delay introduced by the peripheral (keyboard and mouse...usually small but it adds up) + processing time for the display + time it takes for a pixel to shift it's color. That's why most console players don't notice/don't care that their ps4 controller has a 15-20ms input lag by itself.
So the 23ms latency doesn't mean anything. And, tbh, until they fix or get rid of DirectSound, PC gaming audio is held back quite a bit. I found this out when I ran a game through Jriver. Despite it being like....100ms behind, the highend was a lot better and a lot clearer. DirectSound, the window's default audio driver, rolls off the high end for some reason.
Edit: Forgot to say, I bought a Beyerdynamic MMX 300 just to fuck around with the mic input of the Hel. It's pretty damn clear, though it didn't take much to drive the mic to clipping. Though, the mic itself is very fucking sensitive. I'm probably returning it...this headphone doesn't fit on my bigass head well and the clamp is more like a torture device. That said, I finally personally confirmed that the MMX300 is not just a DT770...it's more like a closed DT880, even right down to the imaging. It's kind of bizarre.
@PacoTaco I don't really think that's how that all works together...It's not a cumulative effect at all stages (some places, yes, but more like little loopbacks with some interrupts), nor is everything tied to waiting on a new frame by any means.
20ish-ms of audio latency probably isn't a big deal for gaming, nonetheless. That would be a good wireless setup. Double that and you might start feeling like something isn't quite right.
You're right, I poorly explained that. I was, in a long winded way, explaining the cumulative time for the entire system to output a frame of display data makes any small delay in audio inconsequential to the player.
Edit: I removed half this post because it wasn't exactly relevant to anything.
Still should be ~20ms avg at 144Hz, with a fairly average consumer setup, and assuming G-sync + vsync + fps cap, dipping down closer to ~15ms at 240Hz (same conditions). That's just north of no vsync and fps caps. Plus, some of the latest 240Hz panels have made gains in input latency and pixel response time, and there are some keyboards and mice that shave off an appreciable amount of input lag now too.
And now that you can more or less disable pre-rendered frames in GPU drivers, in theory getting sub 10ms end-to-end latency should be possible if you have the surrounding hardware for it.
Only caveat would be if you're actually taxing your GPU and CPU, but that's not really a good way to measure system latency.
Still a moot point, but I don't want someone reading you post and misunderstanding how their PC and peripheral components all work together.
I’ll have to dig it up somewhere on the nvidia subreddit, but the average lag ends up something like 30 Ms on average with ultra low latency mode at 144hz. It’s usually 50ms with Gsync on and vsync on. But that’s also if the GPU isn’t being close to fully utilized at the 138 FPS cap.
As you said, 240hz destroys that number and most serious competitive plays don’t frame cap.
Folks on reddit have a tendency to pretend to be experts and misinterpret what results they see. That's why they get all gung-ho with sites like ASR. They see some figure posting graphs, which they don't fully comprehend, and take it all at face value. Or they might use stuff they can use at home, which doesn't quite have the level of testing accuracy we need to make absolute claims. (Relative, sure.)
There are guys at sites, like Blur Busters, that have done full end-to-end input latency tests. That is, how long does it take for an input to register on the screen?
144Hz, G-sync and vsync on, but with in-game framerate cap, is an average of just over 20ms.
I think Linus tech tips recently did a video where they checked to see how long it took for a monitor to change with a mouse movement. Was around 15ms, wired or good wireless. (I didn't get full test parameters on that one, so may have been uncapped and without a sync of any type.)
And the Nvidia ULL graph they published during the announcement says it was done at 60fps, 4K, 2060, with what seems to be medium-to-high settings...so that would mean the GPU was probably taxed.
I'm more talking about the video than the subreddit's actual input. Nvidia's subreddit has....way too many fanboys on both sides of that war to be taken seriously at times.
Here's the link I'm talking about: . It is set to 144hz though.
The guy was basically checking how long it would take for him to press his mouse and for a shot to be fired in Overwatch, BFV, ect. Like I said though, this doesn't effect any latency related to Audio though. That's not tied to framerate or anything related to displays at all unless the game dev is a bad programmer.
I did find that linus video (). They were using uncapped CS GO with a 144hz Acer of some kind it looks like and were at about 300 fps (a bit lower, but not much.)
Just got my Hel. Really impressed with it, it's beautifully built and looks great.
Sounds noticeably better than the fulla 2/3, seems to have more body, better low end. I like the switches and knobs, makes me feel like I'm firing up a mini nuclear reactor each time I turn it on.
Drives the 58x nicely on low gain at around 10 o'clock. The 6xx in high gain sounds great at 9:30ish. Lots of power to spare. The mic is Crystal clear and loud, that front mic gain is sweet.
One question: is it normal for it to show up in Windows as "Fulla"?. I do have a fulla 2 and had the fulla 3 for a while. Expected it to show as "hel" or something.
They probably use the same CMedia USB identifier for all three. No harm in that
My Windows 10 states "Lautsprecher (2- Schiit Hel)", but I didn't have a Fulla before.
Seems the hel is quite decent, a juiced up fulla. If I dont particularly care for the mic options would a magni modi stack be better value?
Modi magni will still offer more, yes. Doesn't make sense to go Hel unless you are space constrained or want its mic input.
Separate names with a comma.