Googling around, I am still not sure I got it right:


First there was the Umatic, which in retrospect became low-band
Then Sony created High-band, but for 625/50 only. Did Sony give all these machines "BVU" prefix? Or only some of these machines? High band 625/50 means BVU?
Then Sony created SP, both for 525/60 and 625/50. Were any SP machines ever called BVU? Or BVU was used strictly for pre-SP 625/50 high-band machines? No BVU in the US? If there were BVU machines for the US, then they were SP?

Here (https://umatic.palsite.com/bvu950ovi.shtml) is the BVU-950 Editing U-matic incorporates Sony's Superior Performance (SP) technology. According to the specs, it is a PAL deck.

So for PAL, BVU can mean pre-SP high band, or SP ?

I have worked on BVU's. BVU was Sony's high end broadcast line (3/4) where VO was the step down commercial version recorders. VP were the player only units. There were NTSC versions in the BVU line.There were SP versions in the VO line such as VO9850. In Sony's BVU UMatic product line, the "900" series such as BVU950 were SP units. In the commercial line, the "9000" series were SP units.

@djski, I was wondering whether BVU has any relation to the machine being high band or SP, and it seems it does not. It seems that BVU means pro features like ability to connect to a TBC or timecode generator, but in terms of sheer quality it can be the same low band machine.

Please correct me if I am wrong, in the U.S. it was like low band -> BVU low band -> SP, that is, BVU label was dropped after SP came out.

while in PAL regions it was low band -> high band -> SP. So, I am unsure whether PAL regions had both low band BVU and high band BVU?

Another question: how common was Umatic usage in broadcast? I guess smaller stations used it, but what larger stations used? I found out that a Dallas station, for example, used Umatic for ENG all the way until 1987, when it switched to Betacam. What about editing? I suppose, local news were edited on Umatic and probably played live on Umatic, while something more important was bumped to 1-inch?

UMatic back then was commonly used for commercial insertion and corporate productions.. The broadcast units also had built in time base correction. The term "broadcast" was mostly just a name to designate the higher end product line with added features..
The field guys sure earned there money back then toting big cameras, heavy 3/4 portable recorders and large heavy batteries to make it all work.

Early 3/4" decks were used on air but the station usually got a fine for broadcasting "non-program" material. The translation. The decks had horrible blanking timing errors resulting in a black stripe on the leading left edge. The stripe was the non-program material. The picture did not fill the screen. Early TBC's only fixed the mechanical jitter in any helical format. The BVU series was pretty good on blanking and the onboard TBC helped a lot. My BVU950 retired from an Ohio station is in great shape. And I still have an old school tech that can fix it.

So, BVU means just more connectors and broadcast-friendly features, it does not mean they were necessarily high-band? When SP were released, they were just called SP, not BVU SP or something?

Did stations edit Umatic on Umatic decks, or bumped to 1-inch for editing?

Everything you need to know or answer is here;

https://en.wikipedia.org/wiki/U-matic

Everything you need to know or answer is here;

https://en.wikipedia.org/wiki/U-matic
I appreciate the link, but If I could clearly infer it from wikipedia, I would not be asking here.

So, BVU means just more connectors and broadcast-friendly features, it does not mean they were necessarily high-band? When SP were released, they were just called SP, not BVU SP or something?

Did stations edit Umatic on Umatic decks, or bumped to 1-inch for editing?


Usually stations edited UMatic on UMatic decks, with editing controllers. BVU usually is high band in original UMatic terms.

Having worked with the U-Matic format for the better part of 35 years I will try to answer your questions however despite my experience I'm not an expert.
1st generation VO series, 2600 was the base recorder VP the playback only, both are what's considered lowband, carrier deviation was 1.6 MHz with the chroma downshift to 688 khz resulting in a luma resolution of 250 horizontal lines and about about 30 lines of color resolution. There was an earlier"home" deck with a tuner however I'm leaving this one out. This was replaced in the early 80's with the second generation VO and BVU series of front loading machines. The luma carrier was bumped up slightly but the deviation remained the same resulting in less crosstalk and 3db bump in S/N. The BVU series also has a rear panel DUB connection that allows editing without the up and down conversions of the baseband video signal, the highband was still good for 250 lines and was far more stable. My BVU-870 is an SP with the onboard VITC TCG, this was an option for the BVU and I believe the VO-9800 series. SP pushed the luma carrier to some 6.6 MHz with a 1.5 MHz deviation making it backwards compatible with the previous generation however I have not tried an SP tape on a gen-1 deck. The SP retained the same 688 khz color downshift and the somewhat blotchy color resolution of about 30 lines, luma resolution was about 330 lines and had a better S/N. The BVU and higher level VO editors have flying erase heads for frame accurate editing, lower end editors used brute force overwriting.

3/4 was used for ENG by the big -3 networks through the 1980's, by the end of the 70's many stations had the better TBC's from the likes of For-A, Microtime and CVS so they could use the 3/4 format on air until Betacam became the defacto standard. One of the local stations ran Sony M3a cameras on Sony VO-8800 U-Matic S decks until the early 90's then went to the BVW-25 before becoming a FOX affiliate. In 1994 I was involved with the production of Midwest Country and that was all done on JVC 8250's with JVC KY-2700 cameras. Not the HiDef of today but it's still good overall broadcast quality.

I don't know how much was edited on the 1" type-c, that was more of a prime format used for tape delay and time shifting, my first experience with the format was January 28, 1986 and I still have the original tape. Raw NASA of the shuttle disaster, no narrator or commentary. I know type -c is good enough to feed a broadcast xmitter raw with only the accy TBC, 3/4 needs a bit more correction before it's air legal.

The SP decks started with the BVU-870 for the editor and BVU-150 for the portables, the 950 was the final word of the format and they're still commanding the dollars. I have the 870 and it looks just as good as Betacam SP however I can see the chroma limits on the finer details, saturated colors will bleed over the luma contour borders.

FIrst generation Umatic was 1800 series, before the 2600 series. Here in PAL land we had the VO-1810. https://archive.org/details/VO18101

Thanks for the info, I lumped the 1800 series in the same top loading lowband generation as the 2600/2800 series. IIRC the 1800 was all discreet and electromechanical control while the 2000 series was more integrated circuit and logic based, both are well built and in my opinion the 2600 has better audio than the later decks. A 2 lbs belt driven flywheel runs smoother than the direct drive capstan of the front loading machines

@ARC Tech-109, thanks for the info!


So, in the US, BVU meant just pro connectors and maybe a slight bump in luma carrier, but otherwise not much different from the original low-band. Then when SP came out, it was "SP" without "BVU", for example VO-9800 is just Umatic-SP.
But in Europe the introduction of BVU coincided with high-band? If not, did high-band machines have a special label? Googling... Ah, it seems that high band machines were labeled "Umatic H". So, in Europe BVUs could be either a regular low-band, or "H" or "SP".

I want to make a timeline to make it clearer.

Regarding its usage by stations, here is a Texas KXAS-NBC 5 News Collection (https://texashistory.unt.edu/search/?q=&t=fulltext&q1=&t1=dc_title&q2=&t2=dc_description&q3=&t3=untl_agent&q4=&t4=dc_publisher&q5=&t5=dc_subject&q6=&t6=dc_coverage&q7=&t7=dc_identifier&t8=page&fq=untl_collection%3AKXAS&fq=dc_type%3Avideo&sort=date_a&searchType=advanced&start=15162). Incredibly, they do not allow to search by media type or by camera type or other technical stuff, but they display the media type per clip. So, starting from late 1976 through 1986 all of their news clips were Umatic. The link above shows the last 16mm clip about a robbery, and the next one, with the Carter and Ford debate, is the first Umatic clip, at least the first in this collection.

Here is the direct link (https://texashistory.unt.edu/ark:/67531/metadc1482806/m1/) to the Carter and Ford debate. It is soft, and really no better than standard VHS. it seems that they did not use a TBC when they were digitizing it, considering the skewing on the top. Also, I thought that on Umatic, head-switching noise is not visible. But color-wise it looks ok.

This one (https://texashistory.unt.edu/ark:/67531/metadc1128158/m1/) has horrible green cast on the edges of objects, and of course other colors spill over too, like red.

For comparison, one of the last 16mm clips (https://texashistory.unt.edu/ark:/67531/metadc674002/m1/). "You like to be an object?" - Huh.

US spec BVU was geared more towards integration to the broadcast infrastructure than the VO series, transformer balanced XLR's for the audio external hard genlock, framing servo and the external TBC rear panel connector for the optional Sony BVT time base corrector. There were some improvements with the chroma/luma separation and bandpass filters but the biggest thing was the BVU is far more mechanically robust than the VO series. Dynamic tracking and tape tension, the switching power supply and additional shielding really set the BVU apart. The decks were far more service man friendly despite the addl weight.

Regardless of the baseband the 3/4 U-Matic is still a heterodyne composite format and is limited by the chroma/luma bandpass filters. With the chroma down to 688 khz it's only about 100 khz wide giving it a resolution of about 30 horizontal lines. The luma passband is cut off around 3.08 MHz giving it an academic resolution of about 250 horizontal lines that is similar to VHS Betamax and Video -8. In reality the U-Matic format looks far better than the aforementioned formats due to the higher head writing velocity, linear tape speed and overall stability of that flywheel head drum.
I can't speak to the difference between NTSC and PAL beyond my own experience of watching the BBC in its native format and my eyes feeling like two piss holes in the sand from the slower vertical refresh of 25hz. There seems to be a debate regarding the move to highband in the PAL format, this is something you have to research on your own.
To conclude the BVU is far away from the VO series in the design and overall stability, this is also reflected by the position of the decimal point of the price tag. When a few points of a stations ratings are influenced by the big picture quality and millions are at stake the BVU is a wise investment.

US spec BVU was geared more towards integration to the broadcast infrastructure than the VO series
Right, thanks, I got this.
Regardless of the baseband the 3/4 U-Matic is still a heterodyne composite format and is limited by the chroma/luma bandpass filters. With the chroma down to 688 khz it's only about 100 khz wide giving it a resolution of about 30 horizontal lines. The luma passband is cut off around 3.08 MHz giving it an academic resolution of about 250 horizontal lines that is similar to VHS Betamax and Video -8. In reality the U-Matic format looks far better than the aforementioned formats due to the higher head writing velocity, linear tape speed and overall stability of that flywheel head drum.
Um, this is what I wanted to talk about. I understand that Umatic was preferred to, say, VHS or Beta because of the pro connectors and overall integration of the pro workflow, but if we take the first generation video, it did not look better than consumer formats, and the videos that I linked support this claim.

As for stability, again, the first generation VHS looks good enough, and for further generations a TBC can be used just like for Umatic. I can see that linear Umatic sound is much better than VHS/Beta because of faster tape speed. I suppose that SVideo connector and EDIT mode on consumer SVHS/Hi8 machines could achieve similar quality as Umatic over DUB connector even for a second or third generation, am I wrong?

Anyway, I have another question that I posted in this thread (http://www.videokarma.org/showthread.php?t=275301), but I can repeat it here, since I am getting great answers in this thread.

The other day I was reading Umatic SP specs in a late 1980s magazine, and it said: 330 lines in monochrome mode, 240 lines in color mode. Well, I know that Umatic only has about 30-40 lines of chroma resolution, it is as pitiful as any other color-under format. So, what did they mean?

This low-band VO-5800 model shows different values for monochrome and color:

https://i.ibb.co/wRwJF7y/VO-5800-resolution.jpg

The SP VO-9800 shows the same resolution both for monochrome and color:

https://i.ibb.co/Wt3Gf1H/VO-9800-resolution.jpg


Did they mean that without color, the signal is cleaner, there is no interference in the lower part of the luma band, so it actually does provide higher resolution?
Or did they mean that colorful picture LOOKS less sharp to a human eye?
Or something else?
Why the numbers are the same for SP models even in the "conventional" mode (says 250 lines color/monochrome)?

Other manufacturers who made Umatic machines, did they only make low-band ones, never high-band/SP compatible?

Finally, when they talk about VHS/Beta resolution of, say, 250 lines, do they mean in monochrome mode or in color mode? I suppose they mean color mode, as base VHS is similar to base Umatic.

It means the luminance resolution in the different modes. Chroma resolution was not speced.

Apparently the high end machines could have more luma bandwidth in monochrome mode. Since color subcarrier was not present they didn't have to TBC to accurate color phase.

It means the luminance resolution in the different modes. Chroma resolution was not speced.

Apparently the high end machines could have more luma bandwidth in monochrome mode. Since color subcarrier was not present they didn't have to TBC to accurate color phase.
Did you mean to say, "Apparently the low end machines could have more luma bandwidth in monochrome mode"? I would agree with it. So, chroma interferes with luma? And because high-band machines have wider luma bandwidth, chroma interference does not affect luma resolution that much? I am not sure how TBC is related, considering that old machines did not have built-in TBC.

As others have said, all these machines use "colour under". The chroma is separated from the luma, heterodyned on to a much lower frequency carrier, and recorded as an AM signal. The luma is FM, as it always was.

There will always be compromises separating Y and C, especially before comb filters were common. It's also much harder in PAL than NTSC. There will also be overlap between the lower sidebands of the FM Y signal and the AM C signal. The effects of this are minimised by various means.

I'm just amazed the whole thing worked as well as it did.

625 line PAL has a wider luma bandwidth than NTSC so there was more urgency to move to high band in PAL.

As others have said, all these machines use "colour under". The chroma is separated from the luma, heterodyned on to a much lower frequency carrier, and recorded as an AM signal. The luma is FM, as it always was.
Sure. So why is the difference in resolution in monochrome and color mode for low-band machines, but no difference for high-band machines?

Actually it has to do with the switching of the bandpass filters between color and monochrome. If no 3.58 MHz signal is present the filters are left out and the full recording bandwidth is used thus giving the 300+ lines of resolution. When the 3.58 MHz colorburst is detected the bandpass filters are switched in and everything above about 3.08 MHz is diverted to the chroma down conversion thus cutting the resolution down to the upper limits of the filters limiting the resolution to 240-250 horizontal lines, if the luma was to creep into the chroma downshift side it makes some really wierd distortions. The 250 lines for SP monochrome I suspect has to do with the chroma filters and how the signals are processed, I would have to look at the schematic to give a better explanation. It appears that the SP circuits are completely separate from the highband side. It's possible that engineering decided not to switch the filters around during highband color/monochrome being redundant and adding expense. The VO-5800 is a highband with 250 lines color, I've seen both specs published in various places. There was a medical spec VO that was monochrome only with a 330 line resolution. The 250 line limitation is the bandpass filter slope at about 3.08 MHz, the chroma side comes up at this point and is about 1 MHz wide (500 khz either side of 3.58 MHz) or 4.08 MHz. (MHz X 0.8=lines) to simplify. VHS, Betamax and Video-8 are bandwidth limited to about 240-250 lines in color recording, Betamax-1 was on the higher end of this with a faster tape speed, most of the decks would switch the filters around during monochrome to use the full recording bandwidth.


One big advantage of 3/4 over say VHS & SVHS is the video dynamic range, home grade and some of the "prosumer" VHS isn't as linear on the higher end of the luma and tends to knee and compress on the highlights. I don't know if that is design feature or flaw but it's definitely noticable. U-Matic was less about the cost compromise and focused on the professional who wasn't going to compromise on quality. Panasonic had some "pro" decks like the AG-7500 that have the higher end connectors (BNC video) but the lack of format stability made VHS unsuitable for prime broadcast. You can TBC to a certain extent but the mechanical variations once recorded are going to remain. I'm not going to trust a magazine for accuracy regardless of the source, they're not technical people and mistakes are common in advertising. RTFM (read the factory manual)

Let's clarify the color under process. All tape mechanisms produce timing jitter that is too great for a receiver or monitor to follow the chroma frquency and phase changes. Full bandwidth systems like quad 2 inch correct the whole baseband signal including luma and chroma.
Color under systems separate the hgh frequency part of the composite signal including chroma and luma but the only correction is the phase of the chroma and burst. The jitter of the luma between 0 and 3 MHz is acceptable uncorrected. But no chroma can be allowed to go through the luma channel because it would give a total mess of corrected and uncorrected chroma.

When the signal is known to be monochrome, the color circuits can be turned off and wider bandwidth luma can be passed through the FM luminance channel. Still jittery compared to a machine with TBC, but acceptable.

I think part of the reason early U-Matics didn't have the engineering to provide as good a picture is that the VO series was originally intended to become a consumer format at launch in 1971 and pivoted to be broadcast/prosumer orientated after it failed to get consumer traction.

I have the first U-Matic model VO-1600 as well as the second or third Betamax Sony made (rebadged as a Zenith, but besides cosmetic styling it's no different from it's Sony badged version)....Both machines are basically identical ignoring styling, tape format and the extra EIAJ-8 consumer VTR to TV interface jack on the U-Matic. Both machines have internal RF modulators, TV tuners, and had optional matching external timers for time shifting TV shows. Sony clearly meant the U-Matic to have the same consumer application as it did with the Betamax...It just didn't attract the market Sony wanted so they pivoted it's design to match the customers that were attracted to it and started working on making the Betamax in hopes it would do what U-Matic couldn't in the consumer market.

Let's clarify the color under process. All tape mechanisms produce timing jitter that is too great for a receiver or monitor to follow the chroma frquency and phase changes. Full bandwidth systems like quad 2 inch correct the whole baseband signal including luma and chroma.
Color under systems separate the hgh frequency part of the composite signal including chroma and luma but the only correction is the phase of the chroma and burst. The jitter of the luma between 0 and 3 MHz is acceptable uncorrected. But no chroma can be allowed to go through the luma channel because it would give a total mess of corrected and uncorrected chroma.

When the signal is known to be monochrome, the color circuits can be turned off and wider bandwidth luma can be passed through the FM luminance channel. Still jittery compared to a machine with TBC, but acceptable.

Thanks for the explanation. The colour under signsal from the tape is heterodyned back up to a stable subcarrier otherwise the receiver or monitor wouldn't be able to lock to the burst.

Quite apart from the jitter, a PAL, NTSC or SECAM colour signal has lots of energy at high video frequencies. Even if the FM recording system has enough bandwidth to record high frequency Y, it can struggle with signals that are both high frequency and high amplitude. Hence the use of high band on 2" quadruplex machines, especially in PAL with its higher frequency subcarrier.

I once worked with an A format (predecessor of C format) VTR that had a set of high band boards as an option (Ampex VR7003H). The pictures looked pretty good on high band and you got occasional colour lock. Only a few TBCs could correct a signal like this (Quantel 2001?) and I never had access to one. C format VTRS always had built-in TBCs designed to cope with this sort of signal.

Now we can look back, it's obvious that composite colour systems were a compromise. Quite a serious compromise. But at the time it was the best we could do and they were brilliant achievements.

The comment about high frequency with high amplitude being a problem is correct. RCA's ads for high band quad emphasized they could handle yellow better than the old machines.