CurtPalme.com Home Theater Forum

[Curt Palme]

I took this from the 'curved screen' post that I never read until a customer alerted me to the following statement:

Tom - I totally agree, lens building is still an art, and my experience is
that 3M/Delta has gotten better and better over the years. I have a set
of 2002 HD8's here that are very much nicer than the 1996 models... [i]

I've never done a/b comparisons with lenses, but assuming that the lenses are the same model, are there really differences?

I guess I can also refer to the HD10F lenses on some 1292s, people say that the G90 HD10F lenses are better. Really? Has anyone done some critical comparisons?

Reason being is that I'm selling a set of HD10 to a customer that has HD10L that are wrong for his screen, and he was wondering if later 10s were better than earlier ones.

Just curious..

[Curt Palme]

I took this from the 'curved screen' post that I never read until a customer alerted me to the following statement:

Tom - I totally agree, lens building is still an art, and my experience is
that 3M/Delta has gotten better and better over the years. I have a set
of 2002 HD8's here that are very much nicer than the 1996 models... [i]

I've never done a/b comparisons with lenses, but assuming that the lenses are the same model, are there really differences?

I guess I can also refer to the HD10F lenses on some 1292s, people say that the G90 HD10F lenses are better. Really? Has anyone done some critical comparisons?

Reason being is that I'm selling a set of HD10 to a customer that has HD10L that are wrong for his screen, and he was wondering if later 10s were better than earlier ones.

Just curious..

[Nashou66]

With the HD 145's that are on the longbows I have to say are sharper than the older color filtered set I had.

Not sure the year of the Colored ones but the new Hd-145's are 2004 build.

Athanasios

[ecrabb]

Good discussion, Curt. I've wondered about this myself.

It would explain some of the differences that have been observed with HD-8's... The set of HD-8's I had on my Sony 12xx sucked ass so bad, they wouldn't even begin to resolve 1080i scan lines. The scan lines were obviously observable on the tube face (with a loupe or magnifier), but no way no how could you ever get them on the screen... They were just way, way above the resolution limits of the lenses. Perisoft observed the same issue on the HD-8's on his 808.

Based on my experience, I think most of the reputation of the 12xx being softer, and not that great of an HD machine is due to the lenses, because the other 12xx had the Sony PT-whatever lenses, and it looked identical to the machine HD-8's. I could only conclude based on my observations that the 12xx would be a much sharper machine with better lenses. I never got around to trying because I ended up snagging my G70 for not a lot more money than a set of Joust's adapters and a set of HD-144's.

OTOH, other people say HD-8's are fine, and that 720p and 1080i looks great on their Marquees (which is unbelievable to me based on the set of HD-8's I had.

So, there you go. Obviously, there are some differences. Why 3M/Delta wouldn't have labeled them with revision numbers so that people could sort all this out. All I'm aware of is the old metal HD-8 and the plastic/composite HD-8 rev B (which is what I think I was using). But, those would have been early-90's vintage lenses... Perhaps the late-90's versions that are on 'Quees are newer revisions. Maybe somebody with a late-model 8500 can chime in on labeling or markings on their HD-8's... Or, did any of the later 8500AC's even have HD-8's? Did they switch to 144/145's like Athanasios' Longbows because all HD-8's actually kind of suck for modern graphics/HD resolutions!?!?!?

SC

[Curt Palme]

[Curt Palme]

[Chuchuf]

I believe this to be the case. Maybe Galen will pipe in as he has noticed the differences in the ago of the lens.
This could explain why the HD10F's from G90's are better than the HD10F from the 1292.

Terry

[draganm]

for HD8's, the older ones from like pre 95 have a gold sticker, newer ones have a silver sticker with patent information and a bar code

[Tom.W]

The real improvements came with advanced anti reflective coatings of each element. As for the optimum lens design I doubt that has changed all that much.

A photographic analogy...

http://rick_oleson.tripod.com/index-166.html

[Sparky015]

If you guys read my "wanted, G90 10F thread in the buy/sell section, this topic is discussed. The conclusion is that there is a difference between later year models than previous years. My guess is that the revision numbers/letters are kept between US precision and the PJ manufacturers. No need to market them to us HT folk, as the pj manufacturers are their true bread and butter customers.

[ecrabb]

[ecrabb]

[papalek]

The discusion in the for sale section was very revealing. The newer lenses are better but how much of a difference you will see depends greatly on the abilities of the PJ. At least for the higher end lenses.

[zGman]

Copied from the rapidly disapearing wtb thread - hope sparky got some lenses...we
sure kept it on top for a while.....!

***********

US Precision Optics / 3M has improved their products' performance
over the years without making any model # changes. You can see this
pretty quickly by comparing an old HD8 (say 1994 model) to one
from 2002. Focus and clarity are much improved. Typically the
lens will have a sticker with the model # and a humpteen digit
serial #. I have found the first 4 digits of the serial # correspond
very closely to the date of manufacture of the PJ they are on,
and have assumed this is the year of manufacture of the lens.

I have not seen a 1292 HD-10F later than 1996. Typically that
lens would have somewhat better corner focus than a HD10GT17
from the same year. But a 2002 GT17 will be better in several
respects than either of the older models (clarity and focus).

I also believe that they developed lenses in conjunction with
the PJ manufacturers, so that the details, specs and cost points
are considered proprietary knowledge. Certainly there is more
to know about lens performance than resolution and throughput,
which are typically quoted specs. The quality of antireflective (AR)
coating manufacturing process has certainly improved over
the years, and when you consider the number of elements in
a projection lens - each with two surfaces to be coated - it
seems obvious that incremental improvements in this area
can make a large improvement in performance. One aspect
of performance that is difficult for me to quantify is what I would
call "intra-scene" contrast (or 'clarity'), and this is directly related to
having very low amounts of stray reflected light in the lens
assembly. All the latest model lenses I have used/tested show
improvement in the this area - and one of the best are the
Barco HD120 found on late model 1209S/Runco 1101 and
some of the Runco 1200 (909/cine9) These lenses only
work really well on a limited range of screen sizes however, 7
or 8 feet wide, in my experience, and are somewhat harder
to bring in perfect corner focus.

The G90 HD10F have been used sucessfully in a wide range of
installations, including curved 'torus' screens. Corner focus and
focus uniformity are excellent, as is the resolution and clarity.


****************

It is speculation to say "only" the coatings were improved. I merely offered that as a discussion point.
There are multiple lens elements in the assembly, some are acrylic but most are glass. There is a
very large achromatic doublet which uses two lenses of different refractive index to compensate
for the dispersion caused by the variation of refractive index with wavelength. Every part of lens
building, from the special glass alloys to the grinding and finishing and coating process is an art,
craft and science. It is all subject to improvements in process control and manufacturing
innovation - and the folks at USPL are obviously good techs and engineers who get better with time.

As far as the difference in picture quality between what you have and the G9010F's , bear in mind the
G90 has one of the the most complete electronic focus and stig systems of any CRT, including electronic
control of the cpc magnets (even the 909 has regular old timey cpc magnets).

In addition, the G90 came with high resolution tubes. Bear in mind also that the spec of
resolving "12 line pairs per mm" is measured at the tube face, not the screen. This is
equivalent to resolving a human hair. It is pretty darn difficult to get a line that sharp
on a tube face. A well setup G90 creates test patterns that look like a laser drawing.

In other words, the lens can only show what is on the phosphor, so there is a limiting
factor to improvements based on the level of setup and the tubes, video chain etc.

G

[stefuel]

Well that might explain why that late model 3600 I have hanging in the garage does 1080P so well. It does have the later "bar coded" lables on them.
I'm now curious if that 8500 that MP has been providing 1080P 72 Hz screen shots with has the newer HD-8 lenses on it.

[mp20748]

[mp20748]

[macgyver655]

[CZ Eddie]

[deronmoped]

How about the Hubble lens?

I still can not get over how badly they screwed it up.

Deron.

[deronmoped]

Here is something I discovered about lens.

Using all the tube face makes you use much more of the lens face. For the longest time I tried to get correct linearity with the electronic controls. That is till I found out it was the lenses causing the problem. It seems to be that they do not hold the tolerances all that close getting out to the edges of the lenses. What I did to help correct the problem was swap the lenses around and then rotate the lenses in their relationship to the tube. This helped, but linearity was still off a little out towards the left and right edges. Top to bottom linearity was perfect sense I was not using the edges of the lenses there.

It would be nice if these lenses came in matched sets. Should help in convergence when all them are working in unison.

Deron.

[stefuel]

One thing many here did not know or have forgotted about is the lens spec for maximum raster size.
OK, it's 6:30 AM. Who can tell me what the manufacturers recommended maximum raster size is for any HD-10 series lens?

Prepair to be shocked