Should you spend more money on an RTX 4070 gaming
laptop, or is an RTX 4060 going to be enough? I’ve compared both GPUs in 18 games at 1440p and
1080p resolutions to show you all the differences! The higher tier RTX 4070
laptop GPU has 50% more CUDA, tensor and ray tracing cores compared to the
lower tier RTX 4060 - a fairly big difference, but that’s where the differences end. They both
have 8 gigs of VRAM with the same memory bus, memory bandwidth, and memory speed. They both
have the same power range too, but the 4060 can reach higher boost clock speeds, as the same
amount of power gets spread over fewer CUDA cores. Technically both of these laptops can run with
an extra 25 watts with Nvidia’s dynamic boost, but in the real world with an actual game
running, 4070 and 4060 laptop GPUs hit a voltage limit around 100 watts or so.
Which means that we don’t get more FPS after 100 watts or so. This graph shows the 3DMark
Time Spy graphics score at different power levels, and we can see that the performance from
GPUs starts to flatline after about 100 watts. It also shows us that although the 4070
is always performing better than the 4060 when they have the same power level, a 4070 at
65 to 70 watts matches a full powered 4060. Both laptops run their GPUs at full power,
so we’re comparing the best possible 4060 against the best possible 4070. Expect
different results with lower power limits. I’m using XMG’s Neo 16 gaming laptop to do
this testing. They both have the same CPU, same RAM, and same cooler for a
fair apples to apples comparison. And the liquid cooler was connected to
ensure no thermal limits were present. Let’s get straight into the 18 games! Red Dead Redemption 2 was tested with the game’s
benchmark, and this game had one of the smaller differences out of all games tested. The 4070 was
just 16% faster at 1440p and 1080p resolutions. Not really much of a difference when you
consider that the 4070 has 50% more CUDA cores. Microsoft Flight Simulator is
generally a CPU heavy game, but it still had one of the biggest
performance differences out of all games tested. The 4070 was 22% faster than
the 4060 at 1080p, and 24% faster at 1440p. Forza Horizon 5 was 16% faster on
the 4070 laptop at 1080p and 1440p, but there was a bigger difference seen in the
1% lows. These measure the dips in performance, so the higher result with the 4070 represents
a much more stable and consistent experience. Hogwarts Legacy was tested in Hogsmeade, and the 1% lows were actually worse on the 4070
at 1440p in this one. I’m not really sure why, but it was consistent, though the 4070 still
had the lead when it came to average FPS. That wasn’t the only game where this strange
behavior was seen. DOOM Eternal reached a higher 1% low on the 4060 at the lower
1080p resolution, despite the average FPS from the 4070 being 38% higher - the biggest
difference at 1080p out of all games tested. The Witcher 3 was tested with the ray tracing
preset enabled. This game had the biggest difference at 1440p out of all games tested. I
mean, the 4070 was barely running over 30 FPS, but the 4060 was unusable. Enabling DLSS
and frame generation would help though. Dead Space tends to be one of the more GPU
heavy games we test, and the 4070 was 23% faster than the 4060 at 1080p - the third
biggest difference at this resolution out of all games tested. The 4070 wasn’t quite
able to make it to 60 FPS with max settings, but like most of these games, DLSS
is available to increase frame rates. Apex Legends had a below average
improvement with the 4070, and at 1440p was the smallest
difference out of all games tested, with the 4070 just 11% faster than the 4060. The
4060 wasn’t too far off 100 FPS at max settings. Call of Duty Modern Warfare 2 had a bigger
gain with the 4070, with a 24% higher average FPS at the higher 1440p resolution with
all settings maxed out. Still though, the 4060 was able to hit 60 FPS, so it’s
not as if it’s going to be unusable. Alright after 9 games I think you’ve got
the general idea, the 4070 is ahead in most cases. I’ll just quickly skip through
the other 9 games tested on screen now, instead of wasting your time talking through the
results from each individual game. I think it’s important to test a wide selection of games
so that we can get an accurate picture of the average performance differences to make the
fairest possible conclusion. As I always say, more data equals more better. Let’s
look at those average differences next. On average over all 18 games tested, at 1080p
the higher tier RTX 4070 laptop GPU was 20% faster compared to the lower tier RTX 4060
laptop GPU. This graph shows how much faster each game was on the 4070, and as you can
see, results really vary depending on the specific game. DOOM Eternal was the biggest
outlier here, doing way better on the 4070 laptop. At least in average FPS, because as
we saw the 4060 was actually ahead in 1% lows. If we step up to the higher 1440p resolution
where we’re more GPU bound, the RTX 4070 was now 22% faster than the 4060 on average. This
time The Witcher 3 is the biggest outlier in favor of the 4070, but it was also one of the
few games tested with ray tracing, so I suspect that’s the reason. At least at 1440p, because the
difference was normal at 1080p. Honestly, I was expecting to see a bigger difference, as the 4070
has 50% more CUDA cores, but that wasn’t the case. Here’s how frame rates look if we
instead take the average of all 18 games. I think this better allows us
to visually see the overall differences as a quick summary. It lets us easily see
that even over a wide selection of games, the 4070 has a decent performance uplift
over the 4060. The 4060 was still able to offer more than 60 FPS on average at 1440p
max settings though, not a bad result at all. Alright, so full disclosure, the only reason
we have 18 games here instead of the usual 25 is because our RTX 4060 laptop literally died
during testing. This isn’t actually a 4060 Neo 16, I’m just using it for this video.
Our actual 4060 laptop just never turned back on and we didn’t get it
replaced. I really hope the problem is just because we had an early version of
the Neo 16, but I honestly have no idea. Now that said, the data that was collected
from the 4060 version before it stopped working looks to be completely fine.
Because we’ve also tested two other 4060 laptops in all the same games
and the results are about the same. So as a bit of a bonus, I’ve also done the 25
game summary graphs but included the 7 missing games using a different 4060 laptop. Technically
speaking, these results in the lighter color aren’t as comparable, as the 4060 laptop there was
the XMG Focus 16. It has the same CPU and RAM and a full powered RTX 4060, but no water cooling,
so make of that what you will. Regardless, at both 1080p and 1440p, the overall average doesn’t
really change, even with these extra data points. But yeah, this is why we haven’t
been able to fairly compare things like thermals and battery life. As
that was kind of the whole point of why I had two laptops that were the same,
but unfortunately the 4060 one just died. Alright, so the 4070 laptop performs
better in games compared to the 4060, as expected, but what about the price difference? Prices of both will change over time, so refer to the links below the
video for updates and current sales. And if any gaming laptops with RTX 4060 or RTX
4070 graphics do go on sale, we’ll be sure to add them to our gaminglaptop.deals website. We update
that every day to include all of the latest sales, so make sure you check it out regularly
to save money on your next gaming laptop. Alright, I’ve got a few examples. Let’s
start with the XMG Neo 16, as it’s what I’ve used in this video. The 4070 config is
almost 300 Euro more expensive than the 4060, but the 4070 actually ends up being better
value from a cost per frame perspective, because the performance difference is
bigger relative to the price increase. For another example we’ve got the ASUS
Zephyrus G14. The 4060 model goes for $1600, while the 4070 model is $1850, at least
right now without any active sales, so this could of course change. Again, check
the link below the video for current sales. The more expensive 4070 configuration ends up better
from a cost per frame perspective here as well. The ASUS TUF F15 on the other hand is $1100 for
the 4060, and $1400 for the 4070, so $300 more or 27% more money for the 4070. This results in
the 4060 winning in terms of cost per frame, because the 4070 costs more money to
get the average 20% performance boost. Then with HP’s Omen 17, we’re looking at
$280 extra to get the 4070 over the 4060, which results in the 4070 being slightly
better in terms of cost per frame. But the difference isn't that big, so depends
how far you can stretch your budget. If you can afford it, in many cases I think it’s
worth spending the extra money to get the 4070. But that said, the 4060 is still definitely quite capable. Even without features
like FSR, DLSS or frame generation. But at the end of the day, it all
comes down to the price difference. If it gets to the point where you’re spending
way more than 20% more money to get that extra 20% FPS boost with the 4070, it’s
probably not going to be worth it. And although the 4070 looks more
worthwhile compared to the 4060, in many cases I still don’t
think that the 4070 is worth it. Check this video next to find out why it
might be worth spending even more money to go up to the RTX 4080. Spoiler, the
4080 performs way better than the 4070 and often isn’t that much more money.
And it has 12 gigs of VRAM to 8 which will definitely be useful in future.
So I’ll see you in that one next.
