Replying to Comments: Nvidia GeForce RTX 2060 Review Feedback
Replying to Comments: Nvidia GeForce RTX 2060 Review Feedback
2019-01-24
welcome back to harbor unboxed a few
days ago now I finally got our g-force
r-tx 2060 coverage online and we've
already received loads of great comments
I did predict that a reply to comments
episode would likely follow for that
video and well here it is
I did my best to cover the r-tx 2060 and
the entire r-tx series in that video but
as expected there were still quite a few
good questions of things I either
overlooked or really didn't have time to
get into but before I get into the
comments and start the whole replying to
comments thing I do have an embarrassing
confession to make I screwed up at the
cost per frame graph in that video
unfortunately I was completely unaware
of the mistake I'd made and it wasn't
until viewer Klaus emailed me pointed it
out that I was able to correct it so I
really appreciate him taking the time to
point out my mistake
Klaus was also right on to what I'd done
wrong I accidentally took the cost per
frame data from the RX 570 video from a
few days prior and that was based on the
attorney pier results I failed to update
the FPS numbers to the 1440p results so
yeah that's why they turned out to be a
lot better value than there should have
been so I was meant to just copy and
paste the actual value of the cards and
not the cost per frame so yeah this
meant that the gtx 1050 TI the three
gigabyte and 60 gigabyte 1060 along with
the RX 570 and 580 appeared to be much
better value than they really were so
here is the original incorrect cost per
frame graph and well here is how it
should have looked the second I was made
aware of the mistake I updated the text
by article and then pinned a comment
with the correct graph on the video but
sadly I couldn't go back and undo the
damage already done so please accept my
apology for letting this embarrassing
mistake slip by me okay so let's get
into what was by far the most discussed
topic and that was erm there were
hundreds upon hundreds of comments
talking about the six gigabyte VRAM
of the r-tx 2060 claiming that it wasn't
enough and before too long it would
become a major issue
a viewer the mighty wombat noted that
everyone keeps bringing up these secure
byte versus a gigabyte debate but fails
to take into account that they're
comparing GD d r 6 with gddr5 presumably
this is when talking about the gtx 1070
1070 TI and the 1080 for example let's
talk about capacity first it's true that
right now for some title 6 gigabytes is
right on the edge and under certain
conditions and may not be enough though
I'm yet to see a situation at 1440p that
you can't tweak the settings to avoid
flooding the vram buffer and almost
always you can do this with very little
impact to the visual quality so at least
for now 6 gigabytes of vram is fine at
1440p and the benchmark data that I
provided my r-tx 2060 review certainly
supports that we saw really solid 1
percent frame time performance and
pretty much all the titles and the only
time it wasn't solid I would say is more
a driver bug as the six year by 10 60
was ok so yeah suggest that it's not a
VM capacity issue how the 6 cubic buffer
will fare in upcoming tiles is yet to be
seen but at some point it will almost
certainly be an issue for the 2060 but
again we'll just have to monitor things
and see how this situation develops and
you can bet will do exactly that as for
the memory being faster you know G DDR 6
and all that well that's certainly nice
but it doesn't make up for a lack of
capacity and a perfect example of this
would be the fury X it's super fast for
gigabyte vm buffer became a major issue
before too long so I'll admit paying
$350 for a modern graphics card with
just 6 gigabytes of memory is a bit of
an issue and it's something that I
should have discussed more in my
previous video the problem is for now I
can't really fault it and I hate to
guess as to when you'll start to run
into trouble my guess is that you'll
probably be able to manage it for quite
some time without really too much of an
issue again don't really want to put a
frame time on this because I am just
guessing but I would say for at least
the next year possibly - it'll be a
pretty easy thing to manage and a lot of
titles you probably won't have to manage
anything at all but again that's just a
guess
time will tell
okay moving on black widow said
something along the lines of on average
53 percent faster than 1060 with same
amount of VRAM for 40% more expensive
the 1060 was on average 70 percent
faster than the 960 with three times as
much vram while it was only 25% more
expensive so all in all the 2060 is not
a good product the reason why it looks
so good now is because basically saying
the rest of the r-tx series is bad yeah
I get where you're coming from but it's
a completely different market right now
so the GTX 960 that was one of the
examples you gave that had to compete
with the r9 270x 280 and 285 the rjx
2060 on the other hand well it has to
compete with Vega 56 and I think it's
pretty hard to argue that it doesn't the
arch X 2060 makes the 27 you look like a
pretty poor value option but as for the
rest of the r-tx series looking bad as
Black Widow puts it it isn't the reason
why the 2060 looks good it looks good or
at least appealing because it is the
best value option at that price point
moving on I received loads of really
nice comments as we often do from you
guys that's always massively appreciated
and it helps us get through those big
benchmark grinds anyway the comment I
want to point out here was from El I
think was ELLs MO or something like that
I don't actually haven't run over me now
I put the phone down I think it was Al's
mo and I forgot that wrong sorry but I
picked this one out as an example not
because it was anything wrong with the
comment it was quite nice really but
there was something that I want to talk
about here it was something that we
discussed with our patrons in last
week's live stream and it was something
that I thought I needed to bring to the
main channel so after our video
detailing what happened with Nvidia in
regards to the RT X 2060 sample a lot of
people seemed to interpret that as being
evidence of I suppose all day one
coverage being pure shield jobs to keep
these companies happy for either free
samples or advertising dollars or
perhaps both so let me be clear about
this most other tech media outlets that
provide day one content
fair neutral and they aren't shields as
it's often put for example you might not
agree with the gamers Nexus r-tx 2060
review I actually still haven't watched
that one yet so shame on me I'm a little
bit behind on my GN content suppose
that's both a good and a bad thing
depending on how you look at it anyway
Dingle sidetracked there I'm using us
Steve as an example here because well
gamers Nexus do incredibly good work and
it doesn't really matter whether their
contents brought to you under embargo or
if it's delivered at a later date really
doesn't matter with those guys it's
going to be the same unbiased content
every time and well we hold ourselves to
that same standard and I know a lot of
other media outlets do as well so anyway
I just want to make it clear that we
aren't suggesting or we weren't
suggesting that anyone who signs NDA's
and get samples ahead of release or any
less honest than we are no one really
accused us of that certainly no other
tech media did but I just want to make
that clear because we saw a few comments
then we thought okay yeah that that may
be taken the wrong way so just want to
clear that up okay comment here from
tank to g2 says that 399 Vega 64 is
still a decent deal she's fair enough
comment there are a few people that were
a little more agitated that I wasn't
recommending Vega 64 because it is down
to around $400 but I did point that out
my video I just said that the $400 ones
use the reference cooler and yeah not a
big fan of that one but yeah it is a
decent deal to be honest I personally
would rather have the RT X 2060 though
just because it comes with a nice cooler
you can get the zotac gaming r-tx 2060
twin fan that can be had for $350 and it
will almost certainly run cooler and
quieter again I realize there is some
risk associated with that six gigabyte
vram buffer but for $400 I would really
want to be getting my hands on a custom
AIB Vega 64 card something with a good
cooler before I'd be willing to pull the
trigger that's just my personal
preference but yeah that's why I'm not
recommending Vega 64 because I wouldn't
personally buy it myself
for $50 more than the RT X 2060 okay
another con
he wasn't too much of this criticism on
this video though I did see a few
comments about the card that we use to
Vega 64 version and it's come up a lot
in the past basically the comment is can
you please use another Vega the asou
Strix suffers from throttling due to
poor VRMs well-known issue unfair
comparison I think it's actually the
cooling on the very arms and not the
actual power delivery system itself but
yeah it's my understanding that some of
the earlier models had issues with the
thermal pads I believe they weren't
covering the verum entirely the surface
area of the power stages and that would
cause them to overheat cause the card to
throttle and whatnot myasu Saiga 56 and
64 cards have had the thermal pads
replaced and according to the readouts
in hardware info operating well within
they expected parameters I'm also not
seeing any core throttling I should also
just point out it was there in the graph
so I'll point out anyway that myasu
Saiga 64 card kept up really well with
the liquid version so that it is vague a
64 liquid and I would have thought that
would be the ultimate out-of-the-box
benchmark so based on information I
would have to say that the card isn't
throttling it's not an unfair comparison
and again the Vegas 64 liquid results
were included in that video all right
another variant related question here
Matt can you do tweet AIB Vega 56
under-voltage slash overclocked versus
AIB 2060 overclocked a stock Vega is
pretty underwhelming but tweaked it's a
different beast running fast with
significantly lower power draw yeah look
this is something I could do and I
actually have done similar sort of
content in the past if I do do an
updated version it's not going to be a
big benchmark video just be a look at
how they compare with the tweak settings
and a few new newer more demanding
titles with this stuff you've milage it
will vary when under vaulting so some
cards will undervolt really well some
not so well it's really just like
overclocking it's a silicon lottery deal
so to say all cards will undervolt like
champs is a bit misleading or very
misleading and honestly I don't know
what the average result for under
vaulting is alright this neck
comment is from Jack Villar while I
definitely agree with most that the
price of what you get is too high I
strongly disagree that nobody asked for
ray-tracing
nobody in their right mind would say I
don't want better effects what you don't
want is poor performance this is
literally the first time we've ever had
ray tracing in real time which is
impressive for a single card to do in
the first generation of ray tracing
cards okay so this was an issue with my
video that I saw raised by a few people
and think what we have here is a bit of
a misunderstanding
firstly everyone who complained about
this heard the same thing which was me
saying nobody asked for ray tracing
which in wasn't what I said what I said
was as far as I can tell nobody really
wants it now you might say well you're
splitting hairs here that's pretty much
exactly the same thing and well yeah it
kind of is but as is often the case
context is important so what I really
said was this technology so ray tracing
is basically being crammed down the
throats of gamers and as far as I can
tell nobody really wants it at least in
its current form and that's the real
problem
the current hardware implementation
really isn't good enough to be of any
real benefit so I'm not saying nobody
asked for ray tracing and I'm not saying
nobody wants it I'm saying in its
current form there are very few gamers
who want to pay a premium for it now
look if we could turn on dax-ur
and go suffer a small performance impact
for a noticeable difference to the way
the game looked and without any glitches
such as shimmering and all that kind of
stuff then hell yes sign me up for a
tracing but what we've seen so far is
nowhere near nowhere near justifies
enabling DX er so I'm talking about
battlefield 5 obviously sure that's my
opinion but I think given the feedback
of well the vast majority of you and
other reviewers and guys who are into
tech and play games I feel like my
opinion
here really isn't out of line or that
unusual or differs much from the norm
okay I'll sort of summarized this one
I'll throw the whole comment up on the
screen so I don't misrepresent Eric here
there was a few people that did bring up
sort of a similar argument not a high
percent sure why but I made the comment
saying the r-tx 2080 and 1080 TI have
basically the same performance so he's
pretty well correctly quoted me there
but then sort of seems to misinterpret
what I was saying anyway I'll read out
part of the comment now the r-tx 2080
consistently outperforms the 1080i in
the vast majority of games and the few
tiles that the sanity eye wins in are
typically older titles such as GTA 5
furthermore if you compare titles such
as Wolfenstein 2 or adaptive shading is
use the 2080 absolutely crushes the
1080i with driver optimizations and all
that sort of stuff on the horizon
expected 2080 to pull further ahead then
add on the top features like del SS rate
racing lower power consumption and envy
link support it's clear the 2080 is
superior to the 1080i
okay so rewind based on that Eric B is
upset that I said the rjx 2018 and 1080i
had basically the same performance I
think that's fair unless my testing
differs a lot from what others have
found
yeah it's technically not an untrue
statement it's not really even
misleading to say so naturally you would
assume based on what I've said there
that they are very similar on average
they deliver similar performance the
2080s a little bit faster but not a
whole lot faster so yeah not sure
so he's correct in newer games the 2080
is up to 10 to 15% fast but it's not all
new again so we've seen some newer games
where it's actually slower and it's
certainly slower on older games like GTA
5 so yeah again it wasn't really a whole
lot faster and average I think it was 1%
from memory so not a whole lot faster
Eric then goes on about DL SS which
personally I'm still not sold on I know
Tim's not I think a lot of you guys are
still waiting for more games and further
investigation into how good that looks
and if there's any issues with it
I don't think I need to say too much
more about ray-tracing I think I've
covered that one in probably already in
this video but I've certainly covered at
my opinion on it before so far the same
goes for envy link I don't know why guys
make envy link out to be some sort of
special godlike SLI mode I'm not sure
about that one and just finally wrapping
this one up
Eric B says at the end of his comment if
you want to compare the 1080i and 2080
then comment something like the 2080 is
slightly faster than the 1080i would be
more accurate as saying they have the
same performance is a bit disingenuous
but again Eric you've quoted exactly
what I said I didn't say the same I said
basically the same so yeah I never said
that it would be disingenuous to say
that they are the same because they're
not the same a 1% difference across it's
not the same so yeah I said it's
basically the same performance because
on average the 2080 was 1% faster in my
testing so yeah okay thomas has a
question here I'm just gonna read the
start of it just want to focus on there
DL SS part he says please redo the
benchmarks Wendy LSS is a thing yeah
look DL SS isn't a thing that we can
just redo the benchmarks with the games
have to implement support for the
technology and so far we only have one
example of that and it's a pretty bad
example
in fact it's so bad that Tim didn't
really bother revisit he's waiting at
for it to come or waiting for a better
game to support it because he says that
the AAA implementation in Final Fantasy
is pretty garbage so yeah it's by no
means coming to all games but hopefully
for the sake of RTX owners support will
dramatically improve this year
okay Kelly your resolve says MSI
backplate is plastic or metal Tech deal
says it's plastic but it really looks
like metal so I'm confused
well we've got it here it's very cool
it's very cool and uh yeah it's a it's
aluminum or aluminium depending on where
you where you are in the world but yeah
aluminum definitely 100% aluminium
shroud on the top is plastic but yeah
backplate is
aluminium looks pretty snazzy and yeah
moving on okay Rex rat says what was
your reasoning for saying this wouldn't
be worth the upgrade over the 980ti so I
said it wouldn't be worth upgrading from
a 980ti to the r-tx 2060 and then he
just says is just a case of performance
increase not being worth $150 yeah the
980ti is similar to the gtx 1070 in
terms of performance if your overclock
the 980 - I think it's a little bit
faster than a stock gtx 1070 so anyway
what all that means is that an RC X 2060
is going to be maybe 10% faster on
average than a gtx 980ti so spending
$350 us for a 10% bump doesn't really
seem worth it to me ok last comment here
that we're going to look at this one's
kind of a feel-good nice one but I yeah
I just wanted to touch on it there were
plenty of really nice supportive
comments so thank you very much for all
of those no she says honestly you do
great work even if there are points I
disagree with which they're seldom are
you present your arguments well and
they're based on facts
don't let immature behavior of some
bother you
you and Tim both seem well grounded on
your reviews keep up the great work
so yeah big thank you there to know she
and everyone else who did leave positive
comments and feedback and stuff on that
video that's always great and it's it's
always really nice to hear from someone
who can disagree with us on some things
seldom or otherwise but they can
disagree with us but still engage in
discussion and yeah have just have a
positive discussion I guess is what I'm
trying to say that's always really nice
because it's always very disappointing
to me when we come across some comment
where someone said unsubbed it actually
happened once or twice on that video and
then I asked if they're willing to give
a reason for why they deciding to
announce that their unsubscribing hey
both of these people came back and they
they said that they no longer agree with
some of the things I say or a lot of the
things I say and yeah I mean I thought
that was a bit disappointing because
it's always good to have
difference of opinions sometimes you can
agree to disagree but your the the
opinions and the feedback from you guys
has helped shape and refine our reviews
and we don't always agree but sometimes
it changes the way we look at things or
maybe we thought that wasn't the best
way of going about things and that's why
we always spend so much time trying to
interact with you guys in the comments
and it's probably a reason why we've led
to this reply in a comment series as
well yeah always a bit disappointing
when people who don't agree with you
can't even have any kind of conversation
but anyway getting away from that sort
of negative stuff again it's really
great to hear comments like gnocchi's
and there was loads of them so again
thank you to everyone who took the time
showed their support or just left some
sort of positive feedback as always it
is very much appreciated okay well that
about wraps it up the big ones here were
my botching of the cost per frame data
thankfully that didn't impact the
conclusion in any way because if it did
I'd have to pull the video redo it and
reupload it so I guess there's that but
and it also didn't affect the r-tx 20 60
or any of its direct competitors but it
did make the Lauren cards look much more
impressive than our in terms of value
the other big topics were very own
capacity something that I'll be sure to
keep an eye on very interested to test
all the new big triple-a titles that are
coming out this year and see how the RT
X 2060 compares to something like the RT
X 2070 in those tiles and you know very
goo and all that sort of stuff other the
other thing is you know you can trust
day 1 reviews so we didn't mean to
suggest if we did to anyone that that
wasn't the case in the video about
Nvidia and RR TX 2060 was actually on
that topic there is one more thing I
would like to touch on something we also
discussed in the patreon livestream
other takeaway for some people regarding
that Archie X 2060 review I think it was
told something like dear Nvidia screw us
we have the takeaway from that video for
a lot of people were that we were just
upset we didn't get our for example I
feel like it was explained better than
that or at least should have been so
that shouldn't have been the takeaway
but if not if I didn't explain it as
well as I thought I'd
which is certainly possible the core
issue here being that we were concerned
in video was trying to mute our coverage
because they didn't like what we had to
say about the RT x-series and if that
was the case we certainly wouldn't be
the last to get the ax
the next most critically objective media
outlet would now no doubt be the next to
go
so yeah we were kind of hoping to put a
stop to that one before it got too out
of hand
we were also frustrated with the fact
that in video we were frustrated with
how they handled the situation I guess
is just the quickest way to go about
that one
they promised us a sample a sample in
time for a day one coverage that didn't
happen they had a sign NDA's so that
turned out to be pretty well pointless
since we had no information to leak they
then hit us with a generic response when
we asked why other media outlets had
samples for now we're still waiting to
hear back from a video on the issue so
we have no idea if we'll be sampled for
the next product release or I'll have to
go somewhere else so yeah I guess time
will tell on that one but yeah that's
pretty much everything I think I needed
to or wanted to cover in this video I
think I'll end it here before I continue
to ramble on about things because there
has been quite a bit of rambling in this
video a lot more than usual anyway if
you want to see more content from us if
all my rambling this video hasn't put
you off and you you can subscribe now
you can support us on patreon that I'll
give you access to our monthly live
streams where we discuss a whole heap of
stuff over multiple hours and I just
scored chat where we chat with you guys
on a very regular basis anyway that is
going to do it for this one I am yo
Steve thank you very much for watching
and I will see you again next time
We are a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for us to earn fees by linking to Amazon.com and affiliated sites.