00:00 - 00:05
can Photoshop pull through with the
00:01 - 00:09
Firefly model and create a wizard about
00:05 - 00:11
to cast a spell no it creates what looks
00:09 - 00:15
to be beggar asking me for a spell
00:11 - 00:19
instead of casting a spell 1.75 out of
00:15 - 00:20
five no but you nailed it okay good job
00:19 - 00:24
it's October
00:20 - 00:26
2024 and we can generate any image we
00:24 - 00:28
want at the stroke of our fingertips as
00:26 - 00:31
long as we can describe it we can create
00:28 - 00:33
it using artificial intelligence and
00:31 - 00:36
generative models that make images for
00:33 - 00:38
us this is a fantastic time to be alive
00:36 - 00:40
I know a lot of people and how they feel
00:38 - 00:42
about these generative models but the
00:40 - 00:44
fact of the matter is that these things
00:42 - 00:45
are here to stay you can throw a rock
00:44 - 00:47
anywhere you want on the internet and
00:45 - 00:49
it'll land on some type of generative AI
00:47 - 00:51
model so what I want to do today is
00:49 - 00:54
compare four of the generative AI models
00:51 - 00:55
that I have a lot of experience with of
00:54 - 00:58
which I think are probably the most
00:55 - 01:00
popular in this field the first one
00:58 - 01:03
being mid Journey 6.1 which uses some
01:00 - 01:04
type of diffusion model magnific doai
01:03 - 01:06
which many people don't know this is
01:04 - 01:08
pretty new for magnific they used to be
01:06 - 01:10
just an upscaler we'll talk about that
01:08 - 01:13
later but now they can actually create
01:10 - 01:15
images with prompts and chat GPT which
01:13 - 01:18
uses doll e as its base and Adobe
01:15 - 01:20
Photoshop 2025 Firefly for the
01:18 - 01:22
generation model when we look at these
01:20 - 01:24
chat gbt we generally are going to be
01:22 - 01:27
using more for administrative tasks like
01:24 - 01:30
writing and helping us analyze data but
01:27 - 01:32
it can also make images so it is is one
01:30 - 01:34
that we need to look at now I will be
01:32 - 01:36
judging these images on this criteria
01:34 - 01:39
and you are free to judge along with me
01:36 - 01:41
out of five so each category will get a
01:39 - 01:43
one through five marking we're going to
01:41 - 01:45
look at the creativity of the image that
01:43 - 01:47
was created we're going to look at the
01:45 - 01:49
quality of the result that was created
01:47 - 01:52
from The Prompt used the accuracy to the
01:49 - 01:54
prompt and the usability of the asset in
01:52 - 01:56
my workflow or documentation and then
01:54 - 01:59
the final verdict whether it's a thumbs
01:56 - 02:01
up or a thumbs down all of these images
01:59 - 02:03
will use the ex exact same prompt and we
02:01 - 02:05
will judge them based on the criteria
02:03 - 02:08
that you see here there is one rule and
02:05 - 02:11
the Only Rule and that rule is this I
02:08 - 02:14
had to use the very first image that was
02:11 - 02:17
rendered from The Prompt that I entered
02:14 - 02:19
say that 15 times fast and that's the
02:17 - 02:20
Only Rule the Only Rule that I had was
02:19 - 02:22
as soon as I typed that prompt whatever
02:20 - 02:23
image was the first one to come up
02:22 - 02:25
that's what I had to select the only
02:23 - 02:27
exception is mid Journey because you
02:25 - 02:28
technically don't get an image from mid
02:27 - 02:31
Journey until you pick from one of the
02:28 - 02:33
four versions so I picked the the best
02:31 - 02:35
version that worked for Mid Journey all
02:33 - 02:38
the others I went with the first image
02:35 - 02:40
that was created the first prompt this
02:38 - 02:42
is a landscape image a castle in the
02:40 - 02:45
mountains at Sunset with a moat around
02:42 - 02:49
it and a waterfall pouring into the moat
02:45 - 02:52
epic light at Sunrise oh my gosh this is
02:49 - 02:55
magnific doai I was blown away the first
02:52 - 02:57
time I typed a prompt into magnific with
02:55 - 03:00
the results that I received so the
02:57 - 03:02
accuracy of this prompt I'm going to
03:00 - 03:05
give it a five it's got a moat it's got
03:02 - 03:07
a castle it's got mountains it's got
03:05 - 03:10
what looks to be like a sunrise as far
03:07 - 03:12
as the quality of this I could very well
03:10 - 03:14
use this in one of my images and the
03:12 - 03:16
quality is pretty fantastic the
03:14 - 03:17
creativity I think it looks pretty darn
03:16 - 03:20
good I'd be excited about this and the
03:17 - 03:22
usability I'll give it a four because I
03:20 - 03:24
could definitely use this so this gets
03:22 - 03:26
about a four and a half out of five mid
03:24 - 03:28
Journey with the same prompt that is
03:26 - 03:31
some dramatic looking imagery right
03:28 - 03:33
there I would give that A5 for accuracy
03:31 - 03:35
we have the castle we have the mountains
03:33 - 03:37
we have the sunrise we have the moat we
03:35 - 03:40
have the waterfall everything that I
03:37 - 03:42
asked it created on that the quality I'd
03:40 - 03:45
give it a four it's not the best quality
03:42 - 03:47
when we zoom in on things but if I was
03:45 - 03:49
using this for you know just a thumbnail
03:47 - 03:52
or some type of graphic it could be
03:49 - 03:53
usable not the best but it's okay the
03:52 - 03:55
creativity I think it did a pretty darn
03:53 - 03:57
good job I like the composition it made
03:55 - 03:59
there and the usability if it was a
03:57 - 04:00
little bit more detailed I could
03:59 - 04:02
definitely get down with this one but
04:00 - 04:05
we'll give this one a four out of five
04:02 - 04:06
chat GPT accuracy of the prompt it's
04:05 - 04:09
pretty accurate I mean we have the
04:06 - 04:11
mountains we have the castle we have the
04:09 - 04:14
moat I guess but it looks more like it's
04:11 - 04:16
on a river than has a moat around it so
04:14 - 04:17
I don't know we'll give it a four on
04:16 - 04:19
that one creativity they did a pretty
04:17 - 04:22
darn good job making this pretty
04:19 - 04:24
creative and uh for the usability it
04:22 - 04:26
depends on what I'd be using this for
04:24 - 04:28
but because the quality is so low that
04:26 - 04:30
gets rendered from chat gbt I would have
04:28 - 04:32
to do quite a bit of work work to use
04:30 - 04:35
this one so I'm giving the uh usability
04:32 - 04:38
a two that would be a 3.5 out of five
04:35 - 04:40
not too bad for a really quick rendering
04:38 - 04:42
at a chat GPT which is primarily used
04:40 - 04:44
again if we think about it for
04:42 - 04:46
administrative tests and maybe some
04:44 - 04:51
image brainstorming let's see how
04:46 - 04:51
Photoshop 2025 handled this exact same
04:53 - 04:59
prompt oh Firefly you've got a way to go
04:57 - 05:03
well on the accuracy of the prompt there
04:59 - 05:06
is no Castle here there is no moat uh I
05:03 - 05:09
there is one mountain I guess it could
05:06 - 05:12
be sunrise but maybe sunet I don't know
05:09 - 05:14
who knows the quality is not quite there
05:12 - 05:17
it's very low scale not very uh usable
05:14 - 05:19
in my workflow and the
05:17 - 05:20
creativity it's about a three I the only
05:19 - 05:22
reason why I gave it a three is because
05:20 - 05:24
it it put a mountain in there it looks
05:22 - 05:26
like from Iceland instead of giving me a
05:24 - 05:31
castle so I'm going to give this a 1.75
05:26 - 05:34
out of five you nailed it Adobe good job
05:31 - 05:36
all right now the next one a portrait
05:34 - 05:38
The Prompt was a couple's engagement
05:36 - 05:42
photo looking into each other's eyes
05:38 - 05:44
smiling photorealistic happy muted color
05:42 - 05:47
palette exact same prompt used for all
05:44 - 05:49
four Images all right so you know as far
05:47 - 05:51
as the prompt is concerned I I give it a
05:49 - 05:53
five and the reason why I give it a five
05:51 - 05:55
is I know that the ladies kind of
05:53 - 05:57
missing an i in there but we'll get into
05:55 - 06:00
the quality on that one in a second here
05:57 - 06:02
but I give it a five because it's two
06:00 - 06:04
people that look genuinely happy I know
06:02 - 06:06
it's an engagement shoot because the
06:04 - 06:10
left hand is rested on the gentleman's
06:06 - 06:12
chest and she has a ring on her finger
06:10 - 06:14
that looks pretty close to the earring
06:12 - 06:16
that's in his ear but that's okay I'm
06:14 - 06:18
giving this a five for accuracy because
06:16 - 06:20
it has all the elements that I need and
06:18 - 06:22
I could use this in my workflow if I
06:20 - 06:25
needed to give it a five there the only
06:22 - 06:28
thing that I find not so good about this
06:25 - 06:30
is the eye I just can't see what her eye
06:28 - 06:33
looks like it's not the best
06:30 - 06:35
but 4.5 out of five good job magnific
06:33 - 06:38
you're killing it mid Journey how'd you
06:35 - 06:39
do well I gave them a four I gave them a
06:38 - 06:42
journey of four because they aren't
06:39 - 06:43
looking into each other's eyes per se
06:42 - 06:45
and I don't know if it's necessarily an
06:43 - 06:47
engagement shoot because I can't see the
06:45 - 06:50
ring finger however the quality here is
06:47 - 06:53
very good and it would be usable if I
06:50 - 06:55
needed this in something uh quick for my
06:53 - 06:57
whatever it is that I might be working
06:55 - 07:02
on this could work I would have to do
06:57 - 07:05
some work to it but it's usable 4.25 out
07:02 - 07:07
of five chat GPT where you at I give it
07:05 - 07:08
a for in accuracy because they are
07:07 - 07:11
looking at each other that's about the
07:08 - 07:12
only thing it could be an engagement
07:11 - 07:14
shoot because the left hand is on the
07:12 - 07:16
chest typically how we would see an
07:14 - 07:17
engagement shoot but there is no ring so
07:16 - 07:19
maybe they just got engaged and he
07:17 - 07:20
couldn't afford one and therefore
07:19 - 07:22
they're just happy that they're about to
07:20 - 07:24
get married and they'll get a ring later
07:22 - 07:26
uh I give it a two for the quality
07:24 - 07:28
though because I just can't do much with
07:26 - 07:31
this in anything in my workflow and
07:28 - 07:32
creativity wise I mean I like the the
07:31 - 07:34
muted color palette that it gave it but
07:32 - 07:37
that's really why it scored so high in
07:34 - 07:41
creativity 2.75 out of five and you
07:37 - 07:42
nailed it good job chat GPT I was
07:41 - 07:44
actually pretty surprised that photoshop
07:42 - 07:47
did better than chat GPT on this one but
07:44 - 07:49
not by much we've got a four for
07:47 - 07:51
accuracy we definitely see this couple
07:49 - 07:53
staring at each other the what would
07:51 - 07:55
have given it a five they definitely
07:53 - 07:57
look happy too if I saw a ring on a
07:55 - 07:59
finger we would have had a five at least
07:57 - 08:01
for the accuracy to The Prompt now the
07:59 - 08:03
quality it's pretty horrible man his
08:01 - 08:04
T-shirt is somewhere underneath that
08:03 - 08:06
necklace but it's not connected to the
08:04 - 08:08
other side you know at least they rolled
08:06 - 08:09
up the sleeve a little bit and I give
08:08 - 08:11
them a four for creativity because it's
08:09 - 08:14
not our traditional couple that we're
08:11 - 08:16
used to seeing I I like the Exquisite um
08:14 - 08:18
jewelry that they're wearing however I
08:16 - 08:22
can't use this in my workflow so you
08:18 - 08:24
beat chat GPT but not by much the next
08:22 - 08:26
one a still life so we're looking at a
08:24 - 08:29
still life of a camera on a desk
08:26 - 08:30
surrounded by various camera tools I
08:29 - 08:33
want this be photorealistic have a
08:30 - 08:35
vintage feel with warm light and a Kiros
08:33 - 08:36
skiro appeal which would be you know
08:35 - 08:39
kind of like your Da Vinci Style with
08:36 - 08:42
you know deep dark Rich darks so
08:39 - 08:43
magnific wonderful job here I could
08:42 - 08:46
definitely use this now I don't know
08:43 - 08:48
what a kuana or cavana camera is but I
08:46 - 08:50
could very easily carve that out in
08:48 - 08:52
Photoshop get rid of that and this would
08:50 - 08:55
kind of be difficult to tell that even
08:52 - 08:58
an artificial intelligence model Made It
08:55 - 08:59
by first glance I definitely give it a
08:58 - 09:02
five for Quality because most anything
08:59 - 09:05
that's coming out of of magnific these
09:02 - 09:06
days is magnificent it is wonderful and
09:05 - 09:08
four for creativity and five for
09:06 - 09:10
usability cuz I definitely could use
09:08 - 09:12
this 4.75 out of five very well done
09:10 - 09:14
magnific let's see how mid Journey does
09:12 - 09:18
mid Journey with your
09:14 - 09:21
accuracy not really I mean I wanted
09:18 - 09:23
various uh tools and and Camera objects
09:21 - 09:25
to be around it it put things around it
09:23 - 09:26
but I can't tell if they're batteries if
09:25 - 09:28
they're rolled up pieces of paper
09:26 - 09:30
possibly cigarette butt filters I have
09:28 - 09:32
no idea I don't even know what these
09:30 - 09:36
tools are that are around it so I can't
09:32 - 09:38
really give it um a good score for
09:36 - 09:41
accuracy because it also doesn't have
09:38 - 09:43
the warm light I'm looking for or that
09:41 - 09:46
KIRO skiro kind of feel and it isn't
09:43 - 09:48
very photorealistic so sorry let's see
09:46 - 09:51
here what do you got chat GPT I was
09:48 - 09:53
pleasantly surprised that chat GPT uh
09:51 - 09:55
did a pre pretty decent job on this I
09:53 - 09:57
give it a four for the accuracy because
09:55 - 09:59
it did Target all of the things that I
09:57 - 10:01
wanted it to Target and those things
09:59 - 10:04
could be objects that I could see going
10:01 - 10:06
onto that camera so that is well done
10:04 - 10:08
however the quality of it just isn't
10:06 - 10:10
quite there I'm not sure if I could use
10:08 - 10:12
this in my workflow at all and
10:10 - 10:14
photoshop's Firefly model yep failed
10:12 - 10:16
once again I'll give it a three for
10:14 - 10:18
accuracy because it did Mark some of the
10:16 - 10:20
things like the look that I was going
10:18 - 10:22
for but as far as the quality the
10:20 - 10:25
creativity and usability it's just not
10:22 - 10:26
there the fourth test is going to be a
10:25 - 10:28
humpback whale leaping out of the water
10:26 - 10:30
with Glacier mountains in the background
10:28 - 10:34
on a blue BL Sky day with white clouds
10:30 - 10:36
and photo realistic so magnific did a
10:34 - 10:38
magnificent job as it does you're seeing
10:36 - 10:41
that as a staple across the board
10:38 - 10:44
magnific does magnificent all the time
10:41 - 10:46
5444 4.25 out of five I could very
10:44 - 10:48
easily use this let's see what mid
10:46 - 10:50
Journey's got for us with this humpback
10:48 - 10:53
whale I'll give it a five for its
10:50 - 10:55
accuracy uh but as far as the quality I
10:53 - 10:56
wanted it to be photorealistic and and
10:55 - 10:58
the clouds in the background look like
10:56 - 11:00
they're painted yes I could do some
10:58 - 11:01
things to make it not looks so painted
11:00 - 11:03
but it doesn't really look
11:01 - 11:05
photorealistic it's more of a painting
11:03 - 11:07
and all of the renderings that I got in
11:05 - 11:09
that series of four were the same way
11:07 - 11:12
but I still think it looks pretty decent
11:09 - 11:15
and usable so 4.25 out of five and now
11:12 - 11:18
we got chat GPT H that's not bad I mean
11:15 - 11:20
i' give it a four for uh the accuracy
11:18 - 11:21
but the quality be again you can't
11:20 - 11:23
really use this very well because the
11:21 - 11:27
quality is so low it doesn't produce a
11:23 - 11:29
very uh large scale image for you um so
11:27 - 11:31
out of that I give a 2.75 out of 5
11:29 - 11:32
probably wouldn't use it in my workflow
11:31 - 11:36
but it's one of those things that if I
11:32 - 11:38
needed to uh have some type of um
11:36 - 11:40
brainstorming session I could very
11:38 - 11:41
easily use chat GPT and use this and
11:40 - 11:43
find some other elements that would work
11:41 - 11:47
for me elsewhere let's see how Photoshop
11:43 - 11:50
does not too bad Photoshop okay accuracy
11:47 - 11:52
will give you a four creativity and
11:50 - 11:55
quality usability not quite there again
11:52 - 11:57
because fireflies 1024 resolution that
11:55 - 11:59
we're stuck with I probably wouldn't use
11:57 - 12:01
it for anything but I mean it it it's
11:59 - 12:02
much better than it usually is so
12:01 - 12:04
Photoshop you did wonderful there now
12:02 - 12:06
let's go to our fifth prompt with this
12:04 - 12:08
one I wanted to do a prompt that would
12:06 - 12:11
be more for compositing so if you
12:08 - 12:13
develop compositing assets how would it
12:11 - 12:15
create something that would be on a
12:13 - 12:17
black background that would be easy to
12:15 - 12:19
extract so for this prompt I used a
12:17 - 12:22
photorealistic depiction of an old
12:19 - 12:25
wizard wearing a hood and long cloak
12:22 - 12:27
holding a staff with a crystal on top
12:25 - 12:31
about to cast a spell facing the camera
12:27 - 12:33
on a black background dang magnific did
12:31 - 12:34
a wonderful job and even to the point
12:33 - 12:36
where I feel like he's about to cast a
12:34 - 12:40
spell because the staff is on fire up
12:36 - 12:41
there around this beautiful Crystal man
12:40 - 12:44
the only thing that I would not give
12:41 - 12:46
this any points for uh is the fact that
12:44 - 12:47
it would be hard to even want to put
12:46 - 12:49
something behind it as far as a
12:47 - 12:51
composite is concerned because it looks
12:49 - 12:53
so darn good but if I need to put a
12:51 - 12:54
background behind this this would be a
12:53 - 12:56
wonderful image to have in the
12:54 - 12:58
foreground five out of five across the
12:56 - 13:00
board how did you do mid Journey not too
12:58 - 13:03
bad B this is good because this is
13:00 - 13:05
usable I could very easily use this
13:03 - 13:07
image a little bit even easier than the
13:05 - 13:09
image before because with the quality of
13:07 - 13:10
light that it's given me I could do a
13:09 - 13:11
lot more with this in my post-
13:10 - 13:14
production but I'm going to give it a
13:11 - 13:16
4.25 uh fours in the other categories
13:14 - 13:21
well done mid Journey how did you do
13:16 - 13:23
chat GPT okay okay well accuracy you've
13:21 - 13:25
got us an old wizard but there's a lot
13:23 - 13:28
of crystals everywhere he is about to
13:25 - 13:29
cast a spell however it's not on a pure
13:28 - 13:31
black background it would be harder to
13:29 - 13:33
extract this guy and I probably wouldn't
13:31 - 13:36
use it anyway because I'm not a big fan
13:33 - 13:38
of the look uh and the positioning of
13:36 - 13:40
the fingers it's like okay this staff is
13:38 - 13:42
just levitating because you decided to
13:40 - 13:44
cast a spell with your fingers okay so
13:42 - 13:47
that probably wouldn't be very usable
13:44 - 13:49
2.5 out of five sorry uh chat chpt this
13:47 - 13:51
one just didn't hit it can Photoshop
13:49 - 13:56
pull through with the Firefly model and
13:51 - 13:59
create a wizard about to cast a spell no
13:56 - 14:00
it creates what looks to be like a beg I
13:59 - 14:02
mean I guess it didn't specify on male
14:00 - 14:05
or female but it looks like a beggar
14:02 - 14:09
asking me for a spell instead of casting
14:05 - 14:12
a spell 1.75 out of five no but you
14:09 - 14:15
nailed it okay good job now with this
14:12 - 14:17
one I wanted to do uh renderings of an
14:15 - 14:19
illustration so this was a sticker on a
14:17 - 14:22
black background of a construction
14:19 - 14:24
Hammer striking the top of a metal nail
14:22 - 14:27
with the words nailed it underneath you
14:24 - 14:31
guessed it where did I get that sticker
14:27 - 14:33
from well I tried it in magnific and I
14:31 - 14:34
wasn't a huge fan I'd say that it it is
14:34 - 14:39
good strong Contender but I probably
14:37 - 14:40
wouldn't use it because I don't know
14:39 - 14:43
where the hammer begins and the nail
14:40 - 14:46
ends and if it's actually hit a nail or
14:43 - 14:49
if they're stuck together so 3.25 out of
14:46 - 14:52
five mid Journey how'd you do holy crap
14:49 - 14:54
you nailed it it looks awesome the only
14:52 - 14:55
reason why I gave this a four instead of
14:54 - 14:57
a five for the accuracy is because
14:55 - 15:00
there's no nail here so when I made this
14:57 - 15:01
sticker for this presentation I had to
15:00 - 15:04
make my own nail and I had to do some
15:01 - 15:07
fixing up so it couldn't be a 100% five
15:04 - 15:10
across the board but very well done
15:07 - 15:12
nailed it awesome let's see how chat GPT
15:10 - 15:15
did wow this is actually really good so
15:12 - 15:16
chat GPT did a great job the only reason
15:15 - 15:18
why I didn't give this a five even
15:16 - 15:19
though it has all of the criteria is
15:18 - 15:21
it's not on a black background it would
15:19 - 15:25
be a little bit harder for me to extract
15:21 - 15:27
this but I give it almost a five because
15:25 - 15:29
what chat PT is great at is helping you
15:27 - 15:30
brainstorm for these things rather than
15:29 - 15:32
actually creating the most accurate
15:30 - 15:34
thing but for it to create this instead
15:32 - 15:37
of just a brainstorming session for this
15:34 - 15:39
sticker well done chat GPT let's see how
15:41 - 15:49
did firefly come on
15:44 - 15:52
okay okay 2221 1.75 out of five but you
15:49 - 15:56
guessed it you nailed it good job there
15:52 - 15:57
photoshop with firefly man oh man here
15:56 - 16:00
they are across the board This Is How
15:57 - 16:02
They did all six categories tallied up
16:00 - 16:05
and totaled up we've got magnific coming
16:02 - 16:07
in with the blue ribbon at 4.38 across
16:05 - 16:09
all six categories mid Journey coming in
16:07 - 16:13
in second place with
16:09 - 16:15
4.04 here's a little aside I always used
16:13 - 16:17
mid Journey for my image rendering
16:15 - 16:18
before magnific came out with a prompt
16:17 - 16:20
renderer and what I would do is I would
16:18 - 16:22
use mid Journey images and then I would
16:20 - 16:24
bring them into magnific and upscale
16:22 - 16:26
them to get that quality of look that
16:24 - 16:28
magnific has you don't need to now but I
16:26 - 16:31
will show you that in a second here and
16:28 - 16:32
where that comes into play J gbt nice
16:31 - 16:35
try all right we know what you're good
16:32 - 16:37
at you're great at administrative tasks
16:35 - 16:39
coming up with written things but you're
16:37 - 16:41
better as a brainstormer for making
16:39 - 16:44
images rather than actually creating
16:41 - 16:46
decent images and Photoshop you just get
16:44 - 16:48
the nailed it award at 2.21 here's the
16:46 - 16:51
deal I hope somebody from Adobe is
16:48 - 16:53
watching this now they've been shoving
16:51 - 16:56
this generative AI stuff and what they
16:53 - 16:59
can do down our throat for so long and
16:56 - 17:02
the generative AI models inside of
16:59 - 17:04
Photoshop are absolutely atrocious I
17:02 - 17:07
don't use them for anything I don't even
17:04 - 17:09
use them to replace specific areas on my
17:07 - 17:10
image because I don't think it even does
17:09 - 17:12
a good job at that because it can't
17:10 - 17:15
match the noise patterns of everything
17:12 - 17:17
else that is around it they need to stop
17:15 - 17:21
forcing this down our throat until they
17:17 - 17:24
can get the Firefly model up to at least
17:21 - 17:25
chat gpt's level okay and that's even
17:24 - 17:27
saying something I think for the amount
17:25 - 17:29
that they're shoving it down our throat
17:27 - 17:32
they should be at least at the mid
17:29 - 17:34
journey level I digress however you see
17:32 - 17:36
this image that we took here I took this
17:34 - 17:39
image from chat GPT and I brought it
17:36 - 17:41
into magnific to upscale it to see if it
17:39 - 17:43
could make this crap version of these
17:41 - 17:45
people look somewhat decent and look
17:43 - 17:47
what it came out with so in my opinion
17:45 - 17:49
magnific really is the winner across the
17:47 - 17:52
board because it creates phenomenal
17:49 - 17:54
images from prompts it is on the
17:52 - 17:55
expensive side but if you into image
17:54 - 17:59
rendering and image
17:55 - 18:01
upscaling magnific is fantastic ftic
17:59 - 18:03
because not only can It upscale things
18:01 - 18:05
even your own photography not only can
18:03 - 18:07
It upscale things and add phenomenal
18:05 - 18:10
detail to them it can also create its
18:07 - 18:13
own images now from prompts and I think
18:10 - 18:16
it does the best job out of all of them
18:13 - 18:18
that I have seen to date now I know I'm
18:16 - 18:20
going to get some Litany of complaints
18:18 - 18:21
about generative AI models and why I'm
18:20 - 18:24
even showing this but here's the deal
18:21 - 18:26
folks these models are here to stay
18:24 - 18:28
they're going to be around they're only
18:26 - 18:29
going to get better and it's one of
18:28 - 18:31
those things where if you're going to
18:29 - 18:33
keep your friends close you definitely
18:31 - 18:35
want to keep your enemies closer while I
18:33 - 18:38
don't necessarily generate images and
18:35 - 18:40
call them my own I do use these images
18:38 - 18:43
in my education so that I can help
18:40 - 18:45
people learn the tools in Photoshop use
18:43 - 18:46
the comment section below to comment on
18:45 - 18:48
other generators that you use or maybe
18:46 - 18:51
some strategies that you found for
18:48 - 18:53
success with your prompts if you're here
18:51 - 18:56
to complain about generative AI images
18:53 - 18:57
these comments will be deleted I'm sorry
18:56 - 18:59
I'm just sick of hearing them thank you
18:57 - 19:00
very much for taking the time to watch I
18:59 - 19:01
do sincerely appreciate if you haven't
19:00 - 19:03
done so already please consider
19:01 - 19:04
subscribing because I take difficult
19:03 - 19:06
things in Photoshop and make them
19:04 - 19:09
seemingly simple so that you can use
19:06 - 19:09
them in your workflow today
video to article