or Connect
AVS › AVS Forum › Display Devices › 2.35:1 Constant Image Height Chat › Avatar 1.78 vs 2.27 comparison
New Posts  All Forums:Forum Nav:

Avatar 1.78 vs 2.27 comparison - Page 2

post #31 of 328
I wish this movie came out in 2.35. 1.78 is an injustice to a scope system. If this movie was designed for scope - it would have trumped 1.78 easily. Because it appears cameron did a crop of his original intention just for scope, that is the heart of this argument.

In my opinion it is a weak argument as Avatar as a whole was just above average to me. My problem with Avatar is it is pushing the 3D craze.

-ELmO
post #32 of 328
"Pushing 3D" YES it was flat out made for it and done AMAZING too. No over blown effects thrown in just to use the 3D but all the right stuff to use it the way it should be to tell the story. I thought it was done very well. I will just present it in crop Scope in my theater I saw it in 3D Scope and it looked phenomenal nothing missing.
post #33 of 328
Quote:
Originally Posted by Nasty N8 View Post

I will just present it in crop Scope in my theater I saw it in 3D Scope and it looked phenomenal nothing missing.

Again, you will not be able to reproduce the "scope" presentation you saw in the theater by center cropping this Blu-ray.

You will be missing portions of the picture that both 2.39 and 1.78 theatrical presentations showed by cropping this way.
post #34 of 328
What I find amazing and it might not mean much is we got alvin and the chipmunks 2 and it has a preview of Avatar in scope
post #35 of 328
Quote:
Originally Posted by Franin View Post

What I find amazing and it might not mean much is we got alvin and the chipmunks 2 and it has a preview of Avatar in scope

There is also a 2D Scope AVATAR trailer on U-Tube and a 3D version (also Scope) that allows you to select between several different 3D modes. I had fun with the "cross eyed" version using my DLP projector.
post #36 of 328
Very interesting thread.

So the 2.40:1 'scope' extract is essentially 'panned & scanned', but in the vertical domain, from the 1.78:1 aspect ratio?

In that case you're pretty much stuffed, if you want to do a constant crop, to extract a 2.40:1 aspect ratio, no?

Is it not possible to extract, say, a '70mm style' 2.20:1 aspect ratio for you CIH screens, which would be a little more forgiving to any vertical panning and scanning (which you simply can't do on the fly) that was originally needed on the theatrical 2.40:1 extraction?

Even then though, (and even IF you could pan up and down on the fly), you're still pretty much second guessing James Camerons work here...
post #37 of 328
Quote:
Originally Posted by MoFoHo View Post

Very interesting thread.

So the 2.40:1 'scope' extract is essentially 'panned & scanned', but in the vertical domain, from the 1.78:1 aspect ratio?

So seems to be the case here.

Quote:


In that case you're pretty much stuffed, if you want to do a constant crop, to extract a 2.40:1 aspect ratio, no?

Whilst not "his vision", I will say that this will work. Of course nothing will be confirmed until we get our BD copies and see for ourselves.
Quote:


Is it not possible to extract, say, a '70mm style' 2.20:1 aspect ratio for you CIH screens, which would be a little more forgiving to any vertical panning and scanning (which you simply can't do on the fly) that was originally needed on the theatrical 2.40:1 extraction?

Not on my system. It is either Scope or 1.78:1.

Quote:


Even then though, (and even IF you could pan up and down on the fly), you're still pretty much second guessing James Camerons work here...

Well only time will tell if this works or not. If not, then there is always 1.78:1 which is still great on a CIH system.
post #38 of 328
Quote:
Originally Posted by CAVX View Post

Whilst not "his vision", I will say that this will work. Of course nothing will be confirmed until we get our BD copies and see for ourselves.

Yes, it will work if you enjoy cramped awkward framing, inadequate spacing, chopped heads, obvious cropping with gui overlays being eaten into, and not seeing the focal point in certain scenes.

And it has already been confirmed, and you can see for yourself on the previous page.

OAR/IAR (Original/Intended Aspect Ratio): 1.78
MAR (Modified Aspect Ratio): 2.39

Only OAR/IAR have been provided for us on Blu-ray; MAR (the aspect ratio or dimensions in which a film was modified to fit a specific type of screen, as opposed to original aspect ratio) may never see the light of day because this isn't a case of Cameron wanting/intending people to see the 2.39 framing, it's a case of him providing a 2.39 presentation to make the most of certain theaters.

Cropping the OAR down to 2.37 does not equal the MAR.

In the home Cameron doesn't have to deal with theaters pitching a fuss about 1.78, he can provide only exactly what he wants to be seen on Blu-ray; meaning those with smaller 1.78 displays (CIH included) need to get the coffee table out of the way and pull up the couch.
post #39 of 328
Quote:
Originally Posted by LilGator View Post

Yes, it will work if you enjoy cramped awkward framing, inadequate spacing, chopped heads, obvious cropping with gui overlays being eaten into, and not seeing the focal point in certain scenes.

What? Just how much vital information can 12.5% top and 12.5% bottom actually contain at the same time?

Quote:


And it has already been confirmed, and you can see for yourself on the previous page.

Your starting to sound like a duck. I bet your just waiting for someone to post a screen cap so you can post back with "see I told you so!"

Quote:


OAR/IAR (Original/Intended Aspect Ratio): 1.78
MAR (Modified Aspect Ratio): 2.39

Only OAR/IAR have been provided for us on Blu-ray; MAR (the aspect ratio or dimensions in which a film was modified to fit a specific type of screen, as opposed to original aspect ratio) may never see the light of day because this isn't a case of Cameron wanting/intending people to see the 2.39 framing, it's a case of him providing a 2.39 presentation to make the most of certain theaters.

Umm, try MOST Cinemas. IMAX should not even be discussed as it is not conventional cinema, it is a specialty venue that even AVATAR does not fit the screen off at 1.78:1.

The so MAR you refer to is the term used to describe the panning and Scanning process used to convert CinemaScope to 4 x 3 TV.
Quote:


Cropping the OAR down to 2.37 does not equal the MAR.

No it doesn't.

Quote:


In the home Cameron doesn't have to deal with theaters pitching a fuss about 1.78, he can provide only exactly what he wants to be seen on Blu-ray; meaning those with smaller 1.78 displays (CIH included) need to get the coffee table out of the way and pull up the couch.

Speak for yourself. I don't have a coffee table between the screen and the seats. I like the look of the bare floor area and do wish it to be cluttered.
post #40 of 328
Quote:
Originally Posted by CAVX View Post

What? Just how much vital information can 12.5% top and 12.5% bottom actually contain at the same time?

All of the things I mentioned. If it wasn't vital, Cameron wouldn't have pan 'n' scanned!

Quote:
Originally Posted by CAVX View Post

Your starting to sound like a duck. I bet your just waiting for someone to post a screen cap so you can post back with "see I told you so!"

"Duck" must have been a typo. Or an odd Australian expression.

I already posted screen caps.

Quote:
Originally Posted by CAVX View Post

Umm, try MOST Cinemas. IMAX should not even be discussed as it is not conventional cinema, it is a specialty venue that even AVATAR does not fit the screen off at 1.78:1.

Some theaters showed 2.39, some showed 1.78, some showed 1.85? My local CIH theater (non-IMAX) happened to show 1.78. Yours may have been 2.39. That's really exactly the point- it was shot and intended for 1.78. It was then modified to fit 2.39 screens.

Quote:
Originally Posted by CAVX View Post

The so MAR you refer to is the term used to describe the panning and Scanning process used to convert CinemaScope to 4 x 3 TV.

The term MAR can and has been used to describe that process, but it also perfectly describes what happened with Avatar. It just so happens that the process Avatar went through isn't very common.

It was converted from the OAR of 1.78 to a 2.39 shorter screen by the panning and scanning process, thus Modified Aspect Ratio.

Quote:
Originally Posted by CAVX View Post

No it doesn't.

Since OAR and MAR are the only valid presentations of Avatar, and only OAR has been provided on Blu-ray, it follows that only OAR (1.78) is a valid presentation of Avatar in the home.

Quote:
Originally Posted by CAVX View Post

Speak for yourself. I don't have a coffee table between the screen and the seats. I like the look of the bare floor area and do wish it to be cluttered.

I don't either, I was just paraphrasing what Cameron said.
post #41 of 328
Quote:
Originally Posted by LilGator View Post

All of the things I mentioned. If it wasn't vital, Cameron wouldn't have pan 'n' scanned!

He Panned and Scanned all of his other films to "better fit your TV", and quite frankly, I see this move in image capture nothing more than a jump straight to video.

Quote:


"Duck" must have been a typo. Or an odd Australian expression.

A "duck" is a bird that has a wide flat mouth called a "bill". It often makes this repetitive "quacking" noise. The Aussie term for " duck" is given to a cricket player that just got out and scored no runs - IE "out for a duck". Take that however you like it.

Quote:


I already posted screen caps.

And given the source caused the last thread to be deleted, I'd like to see legit screen caps from the BD.


Quote:


Some theaters showed 2.39, some showed 1.78, some showed 1.85? My local CIH theater (non-IMAX) happened to show 1.78. Yours may have been 2.39. That's really exactly the point- it was shot and intended for 1.78. It was then modified to fit 2.39 screens.

If Cameron really wanted the "biggest image possible", why would a CIH cinema project 1.78:1? Some 3D cinemas also showed this in Scope.

Quote:


The term MAR can and has been used to describe that process, but it also perfectly describes what happened with Avatar. It just so happens that the process Avatar went through isn't very common.

No, I believe it is a first and one I would not call a "break through". Gee lets shoot it for video (with video) so we don't have to modify it later on. Is that working smarter, not harder or just being lazy?

Quote:


It was converted from the OAR of 1.78 to a 2.39 shorter screen by the panning and scanning process, thus Modified Aspect Ratio.

quack, quack, quack

Quote:


Since OAR and MAR are the only valid presentations of Avatar, and only OAR has been provided on Blu-ray, it follows that only OAR (1.78) is a valid presentation of Avatar in the home.

I can't honestly see myself watching it in 1.78:1 unless it does not work in Scope. The only thing that may force me to watch this film in 1.78:1 will be the subtitles should they be located too low in the frame and are clipped by the scaling process. Otherwise, it will be watched that way simply to see the creative difference.
post #42 of 328
Just watched the trailer in scope, beautiful!!
post #43 of 328
Quote:
Originally Posted by Franin View Post

Just watched the trailer in scope, beautiful!!

Whether you like the story or not, this film is pure eye and ear candy and would make awesome demo material. Shame the BD is not going to be Scope.
post #44 of 328
Quote:
Originally Posted by CAVX View Post

Whether you like the story or not, this film is pure eye and ear candy and would make awesome demo material. Shame the BD is not going to be Scope.

Have not seen the movie!
post #45 of 328
Quote:
Originally Posted by R Harkness View Post

I always find myself intrigued by this claim.

Ever since 1080p projection became the standard I can't remember the last time (years now) I've been aware of pixels on a projected image, unless I stick my nose up to the screen and look for them.

That includes seeing plenty of LCD, DLP, SXRD and LCOS projectors over the years. And it includes viewing at close distances as well.

I regularly have my image size (zooming) blown up well into (and pushing) the recommended THX/SMPTE viewing distances for 16:9 and 2:35:1 images and I never, ever notice any pixels.

It's not just me:

KelvinS1965 was a zoomer who bought a lens just to see what it was like with the intent of selling it on as he was more than happy with zooming. After comparing the two he much preferred the image with the lens and kept it. His sig reflects that.

Art Sonneborn sits at around SMPTEs closest (2 x image height IIRC, 61 degrees horizontal viewing angle) and he clearly sees pixels with zooming from the front row. He has to move to the back (3rd) row of his theatre before the image becomes acceptable to him. With a lens he can sit in his front row.

Maybe you can't see pixels but you can still see a difference with 1080 and a lens if you sit close enough for it to matter. If it wasn't the case, there would be little or no difference for it to matter, and no one would buy the lenses since we'd all be happy with zooming.

Gary
post #46 of 328
Quote:
Originally Posted by CAVX View Post

Whether you like the story or not, this film is pure eye and ear candy and would make awesome demo material. Shame the BD is not going to be Scope.

Indeed, and logically, since the "scope" version is pure eye candy, the 1.78 BD will be 33% more eye candy.

Facts. Gotta love 'em.

Quote:
Originally Posted by Gary Lightfoot View Post

It's not just me:

KelvinS1965 was a zoomer who bought a lens just to see what it was like with the intent of selling it on as he was more than happy with zooming. After comparing the two he much preferred the image with the lens and kept it. His sig reflects that.

Art Sonneborn sits at around SMPTEs closest (2 x image height IIRC, 61 degrees horizontal viewing angle) and he clearly sees pixels with zooming from the front row. He has to move to the back (3rd) row of his theatre before the image becomes acceptable to him. With a lens he can sit in his front row.

Maybe you can't see pixels but you can still see a difference with 1080 and a lens if you sit close enough for it to matter. If it wasn't the case, there would be little or no difference for it to matter, and no one would buy the lenses since we'd all be happy with zooming.

Gary

And it's also not just Rich that is skeptical of this claim. I project a 159" diagonal 16:9 image, and I have to get far closer than comfortable to see pixel structure.

You can sit well within spec and not be able to see pixel structure. This is a fact.

In the very extreme end of acceptable seating, you have where Art's first row is. For DLP, up to 2.14x height you can resolve pixel structure, past that point, it's not so clear. 2.5x Height and further you are definitely not capable of resolving pixel structure. For LCoS obviously you have to be closer than these numbers to resolve pixels.

Pixel structure is only changed in the vertical realm with an A-lens, so you are only improving one dimension. If you are sitting at 2x image height and can resolve pixel structure, then adding a lens will only reduce this, not eliminate it (at best). You will still have an identical horizontal pixel size.

All of this is really a moot point for anyone sitting 1 screen width or further away.

You'd be surprised at how few people sit at distances anywhere approaching 2x image height.

Use this calculator to plug in your seating and screen size and it will provide you a lot of these distances, as well as the PPD (Pixels Per Degree). Anything under 30 PPD is noticeable, and 35 or greater is definitely not. Again, I believe that's for DLP, so LCoS will be able to sit closer than those ranges and not resolve pixel structure (LCD, further away).

http://home.roadrunner.com/~res18h39/calculator.htm

To sum it up (and you can read the sources linked on that page that back this up scientifically) an A-lens would do nothing for me as far as pixel resolvability goes. I simply cannot, and am nowhere near resolving pixels sitting at SMPTE Reference distance.
post #47 of 328
I know how it all works, I've seen enough of them.

The fact remains that people who do the comparison more often than not prefer the lens method. Art sits at 2 x, and I like to sit around 2.4 or less. I'm not sure where Kelvin sits but he thought he'd be selling his lens on after testing but found he preferred it to zooming.

Gary
post #48 of 328
Pixel Structure... a little defocus can do the same thing as typical A-lens blurring . For me the associated blur from a typical A-Lens outweighs the advantage of its compression (at least of the ones that I have seen, have not yet seen a true high end A-lens).

Jason
post #49 of 328
Quote:
Originally Posted by Gary Lightfoot View Post

The fact remains that people who do the comparison more often than not prefer the lens method.

This is not fact, it is conjecture.

Besides, who wants to admit the thousands of dollars they just dropped on a lens really didn't make that big a difference? There's nothing scientific about this "conclusion", it's typically biased subjectivity.

Here's some more conjecture to go along with yours. More often than not, people choose to sit 1 screen width away or further (Reference SMPTE is 1.25x SW) because they don't feel the image is watchable or comfortable that close, and thus cannot claim to notice a difference in pixel resolvability.

If you sit around 2.4x SH (1SW), then you aren't seeing pixel structure, so you must prefer sitting much closer to 2.0x to justify your A-lens purchase, no?

I'd like to know who has a JVC or Sony paired with an A-lens and feels the lens improves pixel structure resolvability. They'd have to be sitting at or closer than SMPTE spec allows.

Now, how in the heck did we get on this tangent in a thread about a film that will not ever be able to be presented through an A-lens properly?
post #50 of 328
Quote:
Originally Posted by Gary Lightfoot View Post

It's not just me:

KelvinS1965 was a zoomer who bought a lens just to see what it was like with the intent of selling it on as he was more than happy with zooming. After comparing the two he much preferred the image with the lens and kept it. His sig reflects that.

Art Sonneborn sits at around SMPTEs closest (2 x image height IIRC, 61 degrees horizontal viewing angle) and he clearly sees pixels with zooming from the front row. He has to move to the back (3rd) row of his theatre before the image becomes acceptable to him. With a lens he can sit in his front row.

Maybe you can't see pixels but you can still see a difference with 1080 and a lens if you sit close enough for it to matter. If it wasn't the case, there would be little or no difference for it to matter, and no one would buy the lenses since we'd all be happy with zooming.

Gary

Hmm.

I've seen quite a number of CIH set ups using an A-lens at this point and have yet to spot a difference or advantage in terms of pixel visibility vs non
A-lens set ups (like mine).

So IF the advantage (aside from light output) of adding more pixels into the equation is that you remove visible pixelization...it's a moot advantage to me and, it seems, to lots of other folks. I note that quite a number of AVS members have large screen, sit close, but complaints of seeing pixels these days seems exceedingly rare. It's just hard to accept the extra pixels argument being a big issue in most cases. In this day of 1080p projectors, I'm sure I'm hardly alone at shrugging when this issue is touted by CIH enthusiasts.

However, IF somehow the extra pixels in an A-lens set up confer some other visual characteristics, beyond simply decreasing visible screen door, THEN that could be intriguing. I'm not sure how it could, but I keep my mind open until I do a careful A/B of A-lens vs zoom.
post #51 of 328
Quote:
Originally Posted by R Harkness View Post

So IF the advantage (aside from light output) of adding more pixels into the equation is that you remove visible pixelization...it's a moot advantage to me and, it seems, to lots of other folks. I note that quite a number of AVS members have large screen, sit close, but complaints of seeing pixels these days seems exceedingly rare. It's just hard to accept the extra pixels argument being a big issue in most cases. In this day of 1080p projectors, I'm sure I'm hardly alone at shrugging when this issue is touted by CIH enthusiasts.

If a seating distance of 11ft and a largest potential of 125" 1.78 on your screen is accurate, then not only is it a moot advantage, it's also a non-existent one. You will never resolve 1080p pixel structure in your setup, and from what I remember, 125" isn't even the size you typically project?
post #52 of 328
Quote:
Originally Posted by LilGator View Post

This is not fact, it is conjecture.

Besides, who wants to admit the thousands of dollars they just dropped on a lens really didn't make that big a difference? There's nothing scientific about this "conclusion", it's typically biased subjectivity.

It's not conjecture at all, it's a fact. It's how I decided on my lens, as have others.

The point you miss is that these people quite often do a comparison test before buying. Art was able to compare an ISCO before buying it. Not everyone has that opportunity, but some companies do a 14 money back option so you can try the lens out first and send it back for a full refund if it's not what you want. That's what I did with my first lens after borrowing a cheaper lens from a forum member. Like Kelvin I bought an ISCO 'just to see' and ended up keeping that one and selling the other. If it was a case of zooming being better, why did I buy the ISCO and sell the Prismasonic? If zooming was better I would have sold the Prismasonic and not bothered with the ISCO at all.

No conjecture there I'm afraid.

Quote:
Originally Posted by LilGator View Post

Here's some more conjecture to go along with yours. More often than not, people choose to sit 1 screen width away or further (Reference SMPTE is 1.25x SW) because they don't feel the image is watchable or comfortable that close, and thus cannot claim to notice a difference in pixel resolvability.

Definitely conjecture.

If you sit where THX suggest you sit for your HDTV plasma, you will be at 40 degrees or 2.4 image heights distance. That equates to around 53 degrees for scope in a CIH set up in the same seat. It's not unusual for the pros to set up their clients at that kind of distance for 16:9 since that's where they are comfortable, and scope of course is wider. SMPTE closest is 2 x IH and furthest 3 x IH.

Quote:
Originally Posted by LilGator View Post

If you sit around 2.4x SH (1SW), then you aren't seeing pixel structure, so you must prefer sitting much closer to 2.0x to justify your A-lens purchase, no?

No.


Quote:
Originally Posted by LilGator View Post

I'd like to know who has a JVC or Sony paired with an A-lens and feels the lens improves pixel structure resolvability. They'd have to be sitting at or closer than SMPTE spec allows.

KelvinS1965 has a JVC and an ISCO II. He was a zoomer until he bought an ISCO 'just to see' and saw the improvement so kept the lens. He could easily have sold it on but didn't. His sig says it all.

Quote:
Originally Posted by LilGator View Post

Now, how in the heck did we get on this tangent in a thread about a film that will not ever be able to be presented through an A-lens properly?

That's easy too. You can watch it as 16:9 with side masking, with or without the lens in place

Gary
post #53 of 328
Quote:
Originally Posted by R Harkness View Post

Hmm.

I've seen quite a number of CIH set ups using an A-lens at this point and have yet to spot a difference or advantage in terms of pixel visibility vs non
A-lens set ups (like mine).

So IF the advantage (aside from light output) of adding more pixels into the equation is that you remove visible pixelization...it's a moot advantage to me and, it seems, to lots of other folks. I note that quite a number of AVS members have large screen, sit close, but complaints of seeing pixels these days seems exceedingly rare. It's just hard to accept the extra pixels argument being a big issue in most cases. In this day of 1080p projectors, I'm sure I'm hardly alone at shrugging when this issue is touted by CIH enthusiasts.

However, IF somehow the extra pixels in an A-lens set up confer some other visual characteristics, beyond simply decreasing visible screen door, THEN that could be intriguing. I'm not sure how it could, but I keep my mind open until I do a careful A/B of A-lens vs zoom.

Kelvin also thought like you did, and he bought a second user ISCO lens just to see what the fuss was all about, with the intention of selling it on (pretty much what I did about 5 years ago now that I think about it). He found he preferred the image so kept the lens.

I think for many it's a case of unless you do the comparison you can't appreciate the difference they can make. Until I tried one I also thought that perfect pixel to pixel images were the only way to go but found that wasn't necessarily the case after seeing an A lens with good scaling. I also thought that adding a lens would drastically reduce ANSI, and found that not to be the case either. One of those cases where theory and practice differ in expected results.

Gary
post #54 of 328
Quote:
Originally Posted by R Harkness View Post

Hmm.

I've seen quite a number of CIH set ups using an A-lens at this point and have yet to spot a difference or advantage in terms of pixel visibility vs non
A-lens set ups (like mine).

I looked at it both ways with a high quality 3 chip DLP in my room on the same screen. I could make out pixel structure clearly at a distance between my second and last row with zooming. With the ISCO III only occasionally in my front row.

I have the money to get rid of the A lens, or toss it in the garbage for that matter, if I felt that I would prefer the look of the zoomed image.

I think 4K will make the lens unnecessary but 1080 ,at least in my set up, it gives a better result than zooming. The MTF of DLP is higher than LCOS or SXRD and I'm not sure if this isn't part of it.

Art
post #55 of 328
[quote=Art Sonneborn

I have the money to get rid of the A lens, or toss it in the garbage for that matter, if I felt that I would prefer the look of the zoomed image.

Art[/QUOTE]

Once you do that I'll join you with my 3L an I don't have the money...but I trust your judgement.
post #56 of 328
Quote:
Originally Posted by DaGamePimp View Post

Pixel Structure... a little defocus can do the same thing as typical A-lens blurring . For me the associated blur from a typical A-Lens outweighs the advantage of its compression (at least of the ones that I have seen, have not yet seen a true high end A-lens).

Jason

Yes, don't confuse a basic 2 prism VC lens with a true cylindrical HE lens.

If you look at the screen cap I did, you can actually see (monitor pending) inter pixel lines on my AT screen.

LINK and click on the first image. Take note of the word "focus".
post #57 of 328
What is the price of a 4k projector?
post #58 of 328
Thread Starter 
It's interesting, all the threads in this forum seem to lead to the same thing: lens vs zooming.
post #59 of 328
In all the "scope" shots in this thread, the composition looks extremely cramped. 1.78:1 looks much better. The shots were obviously composed for that ratio.

Whether you use a lens or zoom, 1.78 is how this should be viewed. Otherwise, you are not viewing this film in its OAR.

To me, CIH is all about making 2.35:1 the larger of the common aspect ratios. It's not about hacking up a 1.78:1 image to fit a 2.35:1 screen just because you want to see your scope screen filled.

How many of you cry heresy when people zoom films to fit their 4:3 or 16:9 screens? It's exactly the same thing. And don't give me that vertical vs horizontal resolution malarky. That's just self-justifying horsecrap.
post #60 of 328
Quote:
Originally Posted by Franin View Post

What is the price of a 4k projector?

Last year the JVC was about $180K and the Xenon lamp was about $7K.
New Posts  All Forums:Forum Nav:
  Return Home
AVS › AVS Forum › Display Devices › 2.35:1 Constant Image Height Chat › Avatar 1.78 vs 2.27 comparison