I'm looking for a high quality but reasonably priced (less than $200) HDMI -> USB2 (or Firewire 400) capture device, ideally with the ability to quickly pause and later un-pause recording (which I assume would be a software rather than hardware feature).
I'd be using an HDCP stripper with the device to capture from my HD DVR to my 64-bit Win 7 laptop, and I would use the pause/un-pause feature to skip the unwanted parts of the content. I realize I could edit that out later, but I'd vastly prefer to cut it out in real-time so that I don't have to edit the output. I will not be capturing game play.
Based on all the reading I've done of this sub-forum (and thanks enormously to all the contributors!), I want to be able to capture at 1080p (progressive is better than interlaced, correct?) @ 60k bitrate (at least occasionally). But if that ability would cost me more than $200, I'll accept lesser capabilities.
I don't have a problem using different capture software that what comes with the capture device (I'm in the USA).
Any suggestions, please?
+ Reply to Thread
Results 1 to 30 of 32
Someone may prove me wrong but I do not think that USB2 would be fast enough for capture over hdmi.
Blackmagic has a device but that is for USB3
I have never come across capture software that allows you to pause a recording. IMHO even if it were possible it would not be a good idea. Editing out commercials etc. retaining recording quality is quite quick and easy anyway.
Oh. What makes you think that progressive is better than interlaced ?
Most capture software allows you to start and stop recordings manually but you end up with separate files that have to be joined together later. Pause and resume to the same file just isn't going to happen. PC based recording is not like using a VCR.
The Hauppauge HDPVR 2 1512 (or similar gaming edition version) captures HD hdmi. If you use a graph you can pause / continue the capture with no split file (personally done and tested). Capturing 1080p 30fps (max 30 fps output for 1080p) will be only available if the source is 1080p either 30 fps or 60 fps although you can down convert real time but not on the fly change from i to p within the device driver.
Here's what a simple graph might look like depending on the available filters on your system
Last edited by gll99; 15th Jun 2014 at 22:04.There's not much to do but then I can't do much anyway.
Many thanks for your replies, folks! I apologize for the delay in responding.
I'll reply to each post separately, and intersperse my comments...
Poster vhelp provided a USB2 capability chart for various capture devices, which I've excerpted below:AVerMedia - C875 Live Gamer Portable (LGP) HD Game Capture for PC/PS3/Wii U/Xbox360 up to 1080p, 60Mbps, which is a USB2 (not 3) device
Enosoft DV Processor allows precisely that! It's my all-time favorite capture software! Regrettably, it's only designed for Standard Def DV/DVI capturing, for which I use with my Canopus (now Grass Valley) ADVC-55. Thus, I very much doubt that I'd be able to use the Enosoft software for HDMI capture. But it's excellent pause/resume capability makes me ache with longing for that feature as I move to HDMI capturing.
TMPGEnc Video Mastering Works 5, but that always re-encodes the output. I have other video editors, such as Corel VideoStudio Pro X7, Arcsoft TotalMedia Extreme 2, Nero 14 and others (I vastly prefer commercial video editors over freeware/open-source editors because the latter are too often present obscenely over-complicated settings and interfaces), but I'm dissatisfied with all of them except VMW5. That being the case, I vastly prefer being able to simply prevent capturing unwanted material in the first place!
Again, thanks for your reply.
Enosoft DV Processor has long done precisely what I want. That being the case, I thought that I was justified in hoping at least one HDMI capture software app could also do that. But even if not, I'd far rather have to join multiple segments together than have to edit the unwanted content out! Joining segments is ridiculously easy.
Thanks for all your hard work, gll99!
Also, do not expect have many choices for capture software available to you using a HD capture device with a USB interface. A few will only work (or work well) with the software the manufacturer bundles with the device.
So, as I wrote above, it seems to me that the best HDMI capture device under $200 is the AVerMedia - C875 Live Gamer Portable (LGP) HD Game Capture for PC/PS3/Wii U/Xbox360 up to 1080p, 60Mbps (Amazon link). It fits most of the specifications I listed in my OP, but apparently it cannot capture 1080p at 60FPS from an HD DVR, but it will do so at 30FPS. The software that comes with it, RECentral, does not allow quick pause/resume, but it's still an open question whether any other software can do that.
So my question now becomes, am I correct about that? I won't be using it to capture any gaming, only pre-recorded content from an HD DVR, but I've read that as long as you use an HDCP stripper, it doesn't matter whether you capture gaming or DVR output, since "frames is frames". Does anyone know differently? Or does anyone know of a better capture device for my needs that's $200 or less?
Thanks again, usually_quiet.
AVerMedia C285-AC Game Capture HD II: "Game Capture HD II enables 'pause recording', with which users can pause and resume whenever they decide to skip a few footages [sic]"!
I'm told that the AVerMedia C285-AC (which is also a USB2 capture device) isn't quite as good as the AVerMedia C875 Live Gamer Portable (LGP) that I'm leaning towards, but it's now clear that quick-pausing HDMI -> USB2 capture software does indeed exist! While the RECentral software that comes with the AVerMedia C875 doesn't support that ability, other software does support it! It's just a matter of obtaining it...
Last edited by EmmB; 18th Jun 2014 at 23:49. Reason: Obtained new information
[Edit] I don't know of any perfect USB HD capture devices. The Elgato Game Capture HD probably delivers the best picture quality but in spite of being a device that hardware encodes, the bundled capture software it must use (there is nothing else that really works properly with it) is CPU intensive. It also has a tendency to drop frames.
The Hauppauge HD-PVR II won't provide the same picture quality as the Elgato Game Capture HD, but a lot of people seem quite happy with it. The C285-AC-Game-Capture-HD-II doesn't have much of a track record. It could be a piece of crap, for all I know, but I guess if you insist on a device that lets you pause and resume recording, you may have to gamble on it.
Last edited by usually_quiet; 19th Jun 2014 at 00:59.
Just do a quick read but don't miss reading the Hardware section.
Check the program directory of any multimedia program installed on your computer and see if there are any files ending in .ax (DirectShow). Some of these will provide access to hardware driver functions for tuners, capture cards and all sorts of devices installed on your system. Others read input files, split outputs to create multiple output streams, demux or mux video/audio/CC/subs, convert video, resize video, convert audio, write/dump output files, display video etc... All that is needed is for these filters to be linked via pins in a specific order and manner so that they can communicate with each other.
The image of the graph I posted is the type used in a small program provided by MS in their DirectX SDK known as Graphedit (graphedt.exe). Many software programs make use of the filter modules by linking them together via input / output pins to create a program with a nice GUI wrapped around it so the activity happening underneath is invisible to the user. Graphedit is a tool to manually build graphs using drop down lists to select and load the needed filters. The individual filters are graphically represented as picture boxes with pins that you link by drawing lines to connect them. For the programmer it saves time because it allows them to see how things connect up and make sure things work before writing any lines of code. It happens to be very handy for the average Joe too because it allows us to build, save and run graphs multiple times without actually writing a program. GraphStudio is another option.
I'm not suggesting that building your own graphs manually is good for every situation because it is very dependent on the filters installed on your own computer and their ability to cross communicate with others. Even if I gave someone the .GRF file of the the graph I posted it probably wouldn't work if they don't have an HDPVR2 installed or the Cyberlink drivers. It might not even work if they don't have a drive "I:" because that is where I dumped the output file. It's just a picture of a sample type that can be built.
For me what it does prove when I run it is that the Hauppauge hardware capture driver has the capability to pause and continue a capture without generating multiple files. If Hauppauge, Arcsoft or any other developer wanted to they could easily add a pause-continue function to their capture programs. After all they are running filters too with a fancy wrapper program around it.
Last edited by gll99; 19th Jun 2014 at 00:54.There's not much to do but then I can't do much anyway.
usually_quiet, I again thank you for your time and effort in helping me out with this! Please see my new next post to see what I've learned on my own. I'll address some of your comments here first...
Last edited by EmmB; 23rd Jun 2014 at 02:22.
What I've learned: Elgato Game Capture HD vs. AVerMedia C875 Live Gamer Portable
As I indicated earlier in this thread, a great deal of reading here at VideoHelp Forum and elsewhere had me leaning toward the AVerMedia C875 Live Gamer Portable. So a bit prior to my reading this post by usually_quiet, I went ahead and ordered it.
But then I did read that post, and due to a bit of uncertainty and some premature buyer's remorse after reading it (considering his comments about the Elgato), I went ahead and ordered that, too!
They both arrived Saturday, so I had the fortunate ability to directly compare them both.
Both devices were trivially simple to setup, and I was virtually immediately able to capture HD material from my HD DVR. Here are some main comparisons:
PACKAGING: They both were well-packaged, but the cables that came with the Elgato Game Capture HD were vastly superior! The AVerMedia cables were ludicrously short pieces of third-rate junk, for which they have no excuse. Boo!
CAPTURE SOFTWARE: In terms of the capture software, again the overwhelming edge goes to Elgato. Why overwhelming? Because unlike the Elgato software, the pitiful AVerMedia RECentral software doesn't provide ANY monitoring information on the computer doing the capturing while recording whatsoever! No video monitor, no audio monitor, not even recording time! [ETA: Although the lights on the capture device changes to let you know when you're actually recording]. Apparently you can do that if you connect another computer, but that's absurd!! From what I can tell, it doesn't appear to be a matter of CPU cycles either (though I didn't do any significant testing of that yet). So if the Elgato software can display smooth video, audio, and recording time, why can't RECentral? I'd really like to know the answer.
AUTO TIMING: To my sad surprise, neither software has the ability to set recording limits by either time or disk space. In all my years of video capture, I've always had the ability to start a timed recording so that I could go and do other things without having to keep running back to check it. That's a very poor decision on the maker's fault, and it probably represents the unwarranted assumption that only gamers will be using their products. And although I can almost certainly work around that by creating a QuicKey or AutoIt script, not everyone has that ability.
PAUSE/RELEASE, ETC: Sadly, neither piece of software allows pause and release, as usually_quiet and others correctly predicted. However, to me it's even worse than that, since the Elgato software refuses to recognize the AVerMedia LGP device, and RECentral refuses to recognize the Elgato device. Here, though, the AVerMedia LGP has a strong potential edge, in that it does provide a software interface that other capture software can theoretically connect to the device through. That interface was available to all the third-party video capture software I tested so far, such as Corel VideoStudio and ArcSoft ShowBiz commercial tools, and a couple of open-source capture tools. Unfortunately, no video ever appeared (possibly because the device needed some kind of "start" signal), and I couldn't find any way to capture audio via HDMI. Still, these are early days, and I may well be able to overcome these difficulties eventually, and if so, I might find myself with pause/resume some day.
CAPTURE QUALITY: For myself (and I imagine a great number of others), all the forgoing factors only matter when the capture quality of the devices are reasonably close. Are they? Is it true that, as usually_quiet predicted in his previous post, the Elgato Game Capture HD probably delivers the best picture quality?
No! In my opinion, the two devices are NOT comparable in capture quality!
Why not? Although I was challenged in this thread on my notion that progressive video/capture is better than interlaced, that turns out in this case to be absolutely true!
Also, the chart I excerpted earlier tells a hugely important tale. Here it is again; compare the Elgato with the AVerMedia :
Now, it's certainly true that not all the Elgato captured frames showed such horrible tearing / interlace artifacts. Many were pretty much as good as the AVerMedia LGP's. However, not a single AVerMedia frame showed anything like the ugly artifacts the Elgato produced! Let the truth be said: Progressive IS superior to interlaced!
I have no idea if the AVerMedia LGP actually captures progressive (1080p) frames directly or if it de-interlaces the frames internally, but given the striking results, I absolutely do not care. Ten minutes of captured video was roughly 25% larger from the AVerMedia than from the Elgato, but again, I absolutely do not care. All the various weakness of the RECentral software and all the other factors absolutely pale given the...
overwhelming superiority of the AVerMedia C875 Live Gamer Portable over the Elgato Game Capture HD in terms of non-gaming capture quality.
Guess which one I'm going to keep?
(Note: I have absolutely no association with either AVerMedia or Elgato)
Last edited by EmmB; 23rd Jun 2014 at 02:35.
By the way, I experienced not even the tiniest difficulty with the computer I used to perform the captures. Here are the key details of that computer, which isn't particularly powerful, for anyone interested...
Model: Dell Latitude E5420 laptop
CPU: Intel® Core™ i5-2410M CPU @ 2.30GHz (2C 4T 2.7GHz/2.9GHz, 2.7GHz IMC, 2x 256kB L2, 3MB L3)
Graphics: Intel® HD Graphics 3000 (12CU 12SP SM4.1 650MHz/1.2GHz, 64MB DDR3 1.33GHz/2.13GHz 128-bit, Integrated Graphics)
RAM: 6GB physical, 3.17GB available to 32-bit Windows 7 Pro/SP1
What I see is someone who does not know what to do with interlaced video. No surprise there.
If you want a real comparison it is necessary to upload video. A minute will do. Screencaps are not enough for comparisons, unless you are never going to watch the video and will only do screencaps with whatever terrible playback method you are using. The AverMedia device is discarding half the temporal resolution when it deinterlaces, and playback won't be as smooth.
Last edited by usually_quiet; 23rd Jun 2014 at 10:29.
post was edited.
well, there appears to be some discrepancy or missing pertinent detail in your posts.. like the frame rate, dimensions, etc., was used in your last testing scenario. but at this point i guess it doesn't matter anymore.
on the elgato, it appears you did not configure the capture properly because you can set it to record the incoming source in the sources native format, interlace or progressive. look under the setting icon.
videos that include paused sections will glitch in some players. when you pause video (h264 compression) consisting of different end/start gop/idr frames, playback in some softare players will have a slight hesitation. for example, ffmpeg's ffplay will hesitate slightly. and in mpc-be, it will play nearly smooth. codec/software player/faster systems will produce different results. it would be nice if the pause were to include a fade-out to fade-in feature so that it would not be noticeable.
Glitz = things that are glittering or ostentatious
Glitch = unexplained malfunction (especially a minor one).
To add to post #17 If you use a computer to play interlaced video you should be using software that has decent de-interlacing algorithms. The software you used to play and/or capture the Elgato's video weaves fields to output progressive frames for display, which is not what you would normally want to do when watching videos. If you just used a better player with bob deinterlacing turned on, the individual frames would have similar quality to that provided by the Live Gamer Portable, but you would have 60 frames per second rather than a measly 30.
If you present a TV with interlaced video using a regular DVD player, Blu-Ray player, or media player (WD TV Live or similar) as the playback device all but the oldest or crappiest flatscreen TVs will do a reasonable job at deinterlacing. It should only be necessary to deinterlace 1080i video if you are uploading it to a website (the Live Gamer Portable was made with that in mind so impatient adolescent gamers need not be bothered to do the deinterlacing themselves), ...or if you have to re-size and convert the video for playback using some device that cannot accept 1080i video.
Last edited by usually_quiet; 23rd Jun 2014 at 17:53.
I'm no video expert (else why would I come to VideoHelp to ask questions?), but I'm pretty good at evaluating logic. And your logic in this thread very much seems to me to be rather faulty. For example:
You keep illogically insisting, even in the face of compelling evidence to the contrary, that the Elgato Game Capture HD is capable of higher quality captures than the AVerMedia C875 Live Gamer Portable. Beside the visual evidence I've already provided that demonstrates otherwise (regardless of your carping), you keep ignoring these critical facts: (1): According to the chart that vhelp originally posted, the AVerMedia C875 LGP can transfer TWICE as much data in the same time period as the Elgato, and...
(2): As I reported in my previous post, the file size of the video captured by the AVerMedia C875 LGP at it's highest quality setting is considerably larger than that captured by the Elgato. For the two ~ 1 minute video capture files I'm about to post, the Elgato's output file is 274,124 KB, whereas the AVerMedia output file is 386,841 KB. Unless you're trying to argue that all that extra data is nothing but blank or duplicated frames (which based on careful examination they are not), what kind of counter-evidential anti-logic leads you to conclude that the Elgato produces superior quality output than the AVerMedia?
(3): When I examine the output files with MediaInfo, the Elgato's highest-quality output files always shows the "Scan Type" as (ugly) "Interlaced", whereas the highest quality output files from the AVerMedia LGP always shows the "Scan Type" as (beautiful) "Progressive".
Your logic makes no sense to me whatsoever.
Smart Cutter for DV and DVB, which is one of the extremely few .TS file editors that is actually "Frame Accurate" (as were the 1 minute videos I'll soon post). True, I cropped the edges of the frames I posted, since at that point in the video some 4x3 material was present in the otherwise 16x9 content.
Nero, VLC, etc).
And if the AVerMedia LGP is actually "discarding half the temporal resolution", why are the files it produces so much larger and so much smoother and clearer than those produced by the Elgato? I admit that my lack of expertise may well be confusing me, but if it's throwing half of anything away, the results I'm seeing would seem to be utterly impossible!
Again, your logic doesn't make any sense to me. Please consider all the facts I've presented and explain your logic more clearly to me.
One Minute Video Capture Links:
AVerMedia C875 Live Gamer Portable - 1 Minute Capture Link
Elgato Game Capture HD - 1 Minute Capture Link
AVerMedia (highest possible quality settings) Output File MediaInfo
Overall bit rate mode : Variable Overall bit rate : 42.6 Mbps Codec ID : 27 Width : 1920 pixels Height : 1080 pixels Display aspect ratio : 16:9 Frame rate : 29.970 fps Color space : YUV Chroma subsampling : 4:2:0 Bit depth : 8 bits Scan type : Progressive
Overall bit rate mode : Variable Overall bit rate : 30.6 Mbps Codec ID : 27 Bit rate : 28.9 Mbps Maximum bit rate : 30.0 Mbps Width : 1920 pixels Height : 1080 pixels Display aspect ratio : 16:9 Frame rate : 29.970 fps Color space : YUV Chroma subsampling : 4:2:0 Bit depth : 8 bits Scan type : Interlaced
Smart Cutter for DV and DVB. [ETA: By the way, I did all this careful frame cutting and so forth on my very fast 64-bit Windows 7 Pro/SP1 machine, with an overclocked Intel i7-4770K 3.50 GHz processor, 32 GB very fast RAM, and NVidia GeForce GTX 750]
But more to the point, when I play the captured video with ArcSoft TotalMedia Theater 6, Nero, VLC, MPC+HC, etc, the ugly tearing in the Elgato output is quite painfully apparent. The AVerMedia output file is vastly superior in all those players!
Is it merely because I'm a relative newbie that everyone keeps denying the obvious? I don't get it!
Last edited by EmmB; 24th Jun 2014 at 06:57. Reason: Added machine specs
It appears the Elgato gives proper 3:2 interlaced pulldown while the AverMedia records 2:2:2:4 which gives a more stuttery playback.
Last edited by smrpix; 24th Jun 2014 at 07:55.
I don't know much about the Avermedia device but I have an ElGato Game Capture HD. The latter can indeed capture as 1080p. For example, given a 1080p60 source it can capture as 1080p30. Off the top of my head I don't recall if it can capture 1080i30 as 1080p30. But...
If your source is interlaced 1080i30, capturing it as 1080i30 is superior to capturing it as 1080p30. The former will retain more of the detail and fluid motion of the source, the latter will have deinterlacing artifacts and jerky/flickery motion. A decent player (and any HDTV) will deinterlace the 1080i30 capture on playback so you won't see those comb artifacts. You can always deinterace you 1080i30 caps after capturing. You can never restore the fluid motion of your deinterlaced 1080p30 caps.
Your sample images show improper handling of the chroma channels in the interlaced ElGato caps. The colors from the two half-images (fields) have been co-mingled. That is not a problem of the ElGato captures. It is a result of the way you made the sample images -- an example of your lack of knowledge of how to handle interlaced video. That's one reason to upload sample videos -- which you finally did.
The sample caps you provided are of telecined film. They source was film at 24 fps. It was broadcast (at least, what was captured via the HDMI cable) as 1080i video with hard 3:2 pulldown. The Avermedia cap has detinterlaced that to 1080p30 leaving a duplicate frame every 5 frames (ie, 24p has become 30p by duplicating every 4th frame). On playback that shows a little jerk every 1/6 second. The ElGato cap shows "normal" 3:2 pulldown judder when played back at 60 fields per second.
You might be better off capturing real HD material to evaluate the abilities of the two devices. Try capturing some liver sports like soccer of American football. You'll see what a flickery mess 1080p30 delivers.
Last edited by jagabo; 24th Jun 2014 at 08:18.
Ignorance can be only fixed if someone wants to learn. You don't. You acted like you are the expert here (and continue do so) and yet plainly you do not know what you are doing. You come here asking for advice and then treat the rest of us like newbies when we tell you something you don't want to hear.
You shout in your posts by using large fonts and have now thrown a full-blown temper tantrum.
Speaking of logic, don't you ever bother to think? Of course the Avermedia is deinterlacing. If not, how else does it produce progressive video from an interlaced source? ...and it is deinterlacing just one of the two field in each frame (by doubling each line) and throwing the other field away to get 1080p30. The fields represent different points in time, so it is throwing away half the temporal resolution.
BTW Most editors are a piece of crap as players, and you have NEVER noticed that? LOL
It is apparent that you have zero interest in listening to what the people trying to help you have to say. Go find your own answers, because clearly that is what you will do regardless. I, for one, am unwilling to put up with your arrogance and hyper-sensitivity anymore.
Last edited by usually_quiet; 24th Jun 2014 at 10:00.
In quality image Avermedia C875 is superior over Elgato and Hauppage. No news here.
The real problem, its record any source has progressive via HDMI.
Elgato is very good card. Look for great review here: http://www.thethrillness.com/2014/06/elgato-game-capture-hd-review.html
I have a C285 box and record is Good only, not great. But its record progressive and interlaced source correctly.
The pros about C875 is no need PC to use (Its can record direct to card using SD option) and very high quality image over others USB 2cards.
You cannot really believe that 1080i TV broadcasts de-interlaced to 1080p30 by the Avermedia C875 looks better on a TV than than 1080i TV broadcasts de-interlaced to 1080p60 by the TV.
Last edited by usually_quiet; 24th Jun 2014 at 11:22.
Always say that the best card for 1080i is the C027.
Currently the C985 can match the C027 when used Virtualdub. But both cards are PCI-e is not what the OP wants.
The 285 does a great job with interlaced sources but limited in 15Mb only.
By the way, I recently noticed my ElGato Game Capture HD can record from Macrovision protected VHS tapes. Something to keep in mind if you ever need to record commercial tapes.
A last gift to the OP, EmmB, who will undoubtedly complain that I should have posted this in the first place rather than give him/her a richly deserved verbal kick in the pants for his/her rant.
Here are sections from two screenshots made by VLC. Although these are just highly compressed jpegs, they should be enough to prove a point about software players and interlaced video playback