MSI GeForce RTX 2070 Gaming Z Review + Benchmarks


MSI’s GeForce RTX 2070 Gaming Z graphics
card is their top tier 2070 option in their lineup at the moment, so let’s check it
out and find out what you get for the extra money. I’ll be testing gaming at 4K, 1440p,
and 1080p, as well as overclocking, cooling and of course ray tracing to show you how
it performs. Let’s start by taking a look at the card.
Overall it’s got a black and gunmetal grey look, so pretty colour neutral and should
fit in well with most builds. There are two double ball bearing fans on the front, metal
backplate with a nice brushed finish and MSI logo on the back, and 8 and 6 pin power connectors
on top. It’s worth noting the absence of NVlink here too, as the 2070 does not support
SLI. There’s RGB lighting on the front of the
card, but as I can’t mount it vertically it’s a little hard to show on camera. There’s
also some RGB lighting on the top which shines through the MSI logo, and the lighting can
be controlled using MSI’s Mystic Light Sync software. There are around 15 different effects
available, or you can turn them off if you prefer. For the I/O there’s three DisplayPort 1.4
outputs, a single HDMI 2.0b output, and one USB Type-C port. The card isn’t too big, at 307mm by 155mm
by 50mm it should fit in most cases, and mine weighed around 1.45kg. Here are the specs of this particular 2070
card, the boost clock is quite high, 210MHz higher than the reference spec and 90MHz higher
than MSI’s 2070 Armor OC card that I recently reviewed, but we’ll see how much further
we can push it with manual overclocking later. Like other RTX cards the 2070 also makes use
of GDDR6 memory, and there’s 8GB available here. The system that I’m testing with has an
AMD Ryzen 2700X CPU in an MSI X470 Gaming Pro Carbon motherboard running at stock speeds,
along with 16GB of T-Force Night Hawk CL16 memory running at DDR4-3200 in dual channel.
Check the links in the description for details on all of the components as well as for up
to date pricing. First up let’s check out Battlefield 5,
a game that actually supports Nvidia’s new ray tracing, one of the key selling points
of these RTX cards. For comparison I’ve also got results with RTX on and off, and
testing was done in campaign mode rather than multiplayer, as it’s easier to reproduce. Starting at 1080p I’ve got the RTX on results
in green, and we can see that the average frame rates are less than half as much compared
to running with RTX disabled in blue and purple. RTX was dropping the performance noticeably,
however despite this it was still very playable at medium and low settings, but to me I found
ultra without RTX to look nicer than these lower settings with RTX. At 1440p with RTX
on it was still playable at the lower setting levels, with low settings still able to average
above 60 FPS with the 1% low not too far behind the average. Meanwhile with RTX off the game
was still playing fairly smoothly comparatively at ultra settings. 4K with RTX on is a write
off with the 2070 in this game, but still playable without RTX at the lower levels,
granted it really depends on what frame rates you’re targeting, I can happily play with
60 FPS in this game but I’m sure others would want higher. For the rest of the games I’ve only tested
maximum settings at 1080p, 1440p and 4K, rather than at every single setting level like I
usually do. Let me know what you prefer, it takes significantly more time to test all
setting levels at three resolutions, this way I can include more games in the time I’ve
got available. Fortnite was tested using the replay feature,
at 1080p even with epic settings the frame rates were very high, and then still above
100 FPS with 1440p. 4K was playable, but ideally you’d probably want a higher frame rate
than this, which of course could also be improved by lowering the settings. Overwatch was tested by completing the same
test run through the practice range, and we’re seeing very high frame rates at 1080p and
even 1440p despite the settings maxed out on epic, with above 200 FPS average for both.
Even 4K was still playing smoothly, at over 100 FPS on average. CSGO was tested using the Ulletical benchmark,
and was another game that saw very high frame rates even with the settings maxed out, with
almost 200 FPS at 4K, although the 1% lows improve quite a bit in this test with lower
resolutions, and I doubt anyone playing competitively would seriously consider 4K and high settings
anyway. Rainbow Six Siege was tested using the built
in benchmark, I’ve found this test to pretty much always score high FPS on any half decent
hardware, so even with ultra settings at 4K it’s not too surprising that we’re still
able to average above 100 FPS, with much higher frame rates possible at 1440p and 1080p. Shadow of the Tomb Raider was also tested
using the built in benchmark, and at 1440p and 1080p we’re still able to average above
60 FPS in this test with maximum settings, so expect pretty decent improvements at lower
levels. 4K wasn’t great at max settings, but again I’d expect lower settings to probably
be somewhat playable. Assassin’s Creed Odyssey was tested with
the built in benchmark too, and as a fairly resource demanding game we’re only just
able to average 60 FPS here at 1080p with ultra settings, although I don’t think you
really need a high frame rate to play this one, 1440p with good settings should still
go alright. PUBG was tested using the replay feature,
and for an unoptimized game it was still giving pretty nice results at ultra settings at 1080p
and 1440p. Pretty poor performance at 4K though, but it might actually run alright with low
settings, but for this sort of game you’d probably want a lower resolution to maximize
FPS anyway. Far Cry 5 was tested with the built in benchmark,
and at ultra settings the frame rates are quite nice considering we’re using ultra
settings for the 1440p and 1080p tests. Not too terrible for 4K, again considering max
settings, so we could probably get decent FPS with lower settings at 4K. Watch Dogs 2 is a resource intensive game,
however I’ve found that I can play it perfectly fine with a solid 30 FPS, and that’s what
we’ve got here at 4K, although I probably would drop the settings a bit to get a little
higher frame rate, as I don’t think you get much of a visual improvement going from
say very high to ultra. 1440p and 1080p on the other hand ran very nicely, even at ultra
settings. The Witcher 3 was tested with Hairworks disabled,
and even with 4K at ultra settings we’re only just below the 60 FPS sweet spot with
a 1% low result not too far below the average, with some slight setting drops it should play
pretty well at 4K, and no problems at all at 1440p and 1080p. Shadow of War was tested using the built in
benchmark, and fairly good results here too. Considering 4K was averaging 50 FPS in this
test at max settings, it would likely actually be playable with lower settings at 4K. As we’ve seen we’re able to get some pretty
good performance in the games tested, 1440p with good settings ran well for the most part,
while 4K wasn’t a great experience, at least with max settings which I’ve shown here.
In some less demanding games like Overwatch you’d be able to run alright at 4K with
good settings, and for others it may be alright at low settings, but for the most part MSI’s
2070 Gaming Z seems to be aimed towards 1440p gaming with good settings. Ray tracing on the other hand in Battlefield
5 was disappointing, hopefully performance will improve in the future as it’s still
a new technology, but the entry level RTX 2070 is only giving us playable performance
at 1080p and 1440p with medium or low settings, and that’s assuming you’re not after high
frame rates, which are generally beneficial in games like this. For overclocking I’ve retested with Far
Cry 5. I was able to overclock the GPU core by 60MHz and the memory by 400MHz, although
I didn’t spend too long tweaking the memory. This wasn’t that much extra, but keep in
mind it’s already overclocked pretty well out of the box. I’ve retested with ultra settings which
should be more GPU demanding, and at 1080p there’s only a small 0.2% improvement to
the average frame rate with the overclock applied, a larger 4.2% improvement at 1440p,
and then a 5.4% improvement at 4K, better results at higher resolutions as we become
less CPU bound. So we’re seeing some decent performance,
but what are the thermals like? These are the temperatures I measured with the Heaven
benchmark using Hardware Info with an ambient room temperature of 23 degrees Celsius. It
wasn’t getting that hot in my testing, and even with the manual overclock applied it
was only a little warmer at 70 degrees. It was fairly cool at idle, considering the fans
weren’t spinning at all, they don’t even spin up until it reaches 60 degrees Celsius.
With the fan maxed out there was a large drop in thermals, but the fan noise rose quite
a bit too as a result. Just for fun here’s how the backplate looked
after 30 minutes of stress test running, getting to about the 65 degree Celsius mark. I’ve got the average and peak clock speeds
while running the Heaven benchmark for an extended period, and with the overclock applied
I was able to average around 143MHz above the stock speed by increasing the core clock
by 60MHz and increasing the power limit to 111% using MSI Afterburner. These are the average and peak fan speeds
measured under the same tests as the temperatures just shown. Even at stock they were quiet
compared to the rest of the system, making almost no difference while gaming even once
manually overclocked. With the latest version of MSI’s Afterburner I was able to boost
the fans to 100%, reaching over 2600 RPM and finally becoming fairly audible. Here’s
how these tests sounded so you can get an idea for yourself. As I’ve got 7 other fans in the case there’s
not much difference, but at idle the fans on the graphics aren’t spinning at all,
so using that as a baseline there’s almost no difference between idle and the stress
tests. It’s just not making much noise until we manually boost the fans. I’ll also note that while testing there
was no noticeable coil whine with my card. Here’s what total system power draw looked
like in the same Heaven benchmark, so once manually overclocked there’s a 5% increase
to power draw for a 7% improvement to clock speed, which resulted in a 5% improvement
in Far Cry 5 at 4K with ultra settings. MSI recommend a 550 Watt power supply or above,
and note that the card itself has a power consumption of 225 Watts. I was seeing around
a 227 Watt difference with my GPU stress test running, though I’d expect some of this
would be increased CPU activity too. For up to date pricing check the links in
the description, as pricing will change over time. At the time of recording, here in Australia
the MSI GeForce RTX 2070 Gaming Z is going for around $1039 AUD, or just under $600 USD
on Amazon for my international viewers. The Gaming Z appears to be the top end 2070
in MSI’s line up at the moment. In Australia this makes it $100 more expensive than MSI’s
2070 Armor OC card I recently looked at, but a lower $30 difference in the US. The Gaming
Z definitely has a nicer overall build quality, there’s more lighting, and due to the higher
boost clock out of the box it was performing a little better, though this only resulted
in a few frames per second extra, so you’ll have to decide if this more premium 2070 option
is worth it for you. In conclusion MSI’s RTX 2070 Gaming Z graphics
card seems to be well suited to 1080p and 1440p gaming, but if you’re after 4K gaming
or actually getting good frame rates with ray tracing outside of low settings at 1080p,
you may need to look at a 2080 or 2080Ti instead. The card looks great in my opinion, but does
cost a bit more than other 2070’s available, so you’ll have to weigh up the features
and how much performance you can likely gain overclocking a cheaper card yourself. So what did you guys think of MSI’s RTX
2070 Gaming Z graphics card? Let me know your thoughts down in the comments, and leave a
like on the video if you found it useful. Thanks for watching, and don’t forget to
subscribe for more GPU testing, including MSI’s RTX 2080Ti Gaming X Trio which is
coming up next.

78 thoughts on “MSI GeForce RTX 2070 Gaming Z Review + Benchmarks

  1. At the MSRP it's actually a very good buy. It's silent, cool, looks nice and faster than an average gtx 1080. Just don't use it for RTX and 4k gaming (AAA titles), lol. Very solid 1440p and great 1080p ultra card.

  2. That's a really powerful card. But the performance bump isn't that much higher compared to a gtx 1070. Gotta stick with my zephyrus until next gen card comes out.

  3. Hello Jarrod's Tech, I enjoy your usual format. But I respect the change and the fact that you are testing more games. I like the usual format better. RTX 2070 is a great card and I love the look of these new MSi cards

  4. Great review as usual
    Jarrod, i'm just curious. Can you see the differences between highest settings (ie ultra) and 2nd highest settings (ie very high) when playing games? I personally don't see any differences so i usually set the game settings to under the highest settings to reduce the stress on both my cpu & gpu and have more framerates.

  5. The 1080ti is still the best value nVidia card on the market. Shame nVidia stopped making them to force people to buy their overpriced 2080 and 2080ti cards. 🙁

  6. Will u do the other rtx from aorus,zotac,gigabyte,Asus. Some even have a triple fan. Will there be any temperature difference

  7. If you just test one setting I’d use high instead of ultra, just because sometimes the max settings have weird stuff like hairworks that people will generally turn off to be able to play at higher res while still keeping respectable levels of eye candy. Good video man, cool to see reviews now that some RTX functionality is out. 👍

  8. A nice break from the laptops lol! 👍
    Also can you do live gameplay videos for different gaming laptops so we can take in the benchmarks better?

  9. Hi, Jarrod!
    Well, the same comment I made on the Armor OC goes to this one, with the difference the feast is more expensive now.
    Have a great weekend, mate. REST!!

  10. I have bought a cyberpower pc with rtx 2700 with ryzen 2700 in a deal for 1200 USD, is this overkill for 1080p 60hz, my idea is that I do not have to worry about upgrading it for at least 5 years.

  11. Great video, I’m going to build my first PC in few weeks n going to get this card tomorrow so KEEN for it, wish me luck boys

  12. If this /watch?v=QnOCu5GJ9SY&t=1s is true, then AMD and us gamers have a bright future! Nvidia on the other hand went full apple with the pricing of their rtx cards, to compensate for lower sales numbers and please their shareholders.

  13. Im getting really low fps using rtx 2070+ryzen 5 2600+8gb ram+asrock b450m-hdv. For example around 110 fps on overwatch at 1080p without ultra settings. Dunno why.

  14. I AM MSI FANBOY! FYI the 2070 FE allowed me 190 OC max, set to 155 OC for daily. Memory +965 OC with 550 OC daily. SO seems to be some truth about binned being FE and I also hear binned for gigabyte aorus cards. MSI usually always OC best. Guess you got a below average card or this is what to expect from MSI inventory as Nvidia cherry picking?

  15. IMO 4k 60 is attainable with just the 2070 in a lot of titles and the ones that won't quite hit the 60fps mark are single player adventure games. If one were to use NO.. AA and run the game with a mix between high and very high settings NOT FUCKING MAXED AT ULTRA i think this card could pull 4k-60 easily. Ultra doesn't add much visual improvement to justify the use of ultra for the performance hit, Also shadows take a big hit to FPS so i'd use medium settings for those,. who the fuck is standing around in their game analyzing shadows lol, half the time you'll be running right past the cast'd shadows or constantly moving around to even notice them all that much. So to me the 2070 is a fantastic card and a economical no-brain'er, something like the 2080ti is not worth it at it's price and is only for people with deep,deep pockets and like to brag, so save that extra $700 and put it towards other things and get the 2070 at least until the other cards come down in price most likely late 2019 and probably not by that much. To re-elaborate i also think using the settings i mentioned all titles would that desired 4k 60 dream, if you disagree tell me why.

  16. What about the gpu anti sag bracket? Is it worth using it? Does it look nice? Thinking I’d paint mine blue and use it but idk if it’s worth even having it there.

  17. I really like how you tested thermals and fan noise. I have the Gigabyte 2070 OC White and with manual overclock of +60Mhz, +300Mhz memory, I am having pretty high temps around 75C, where the fans are ramping up to around 70-80%. It gets pretty loud so I’m looking into a better thermal performer.

  18. This is EXACTLY the card review video which I have been trying to find! Thank you so much!

    To provide some feedback to your question around 3:20, I am personally fine with only having BFV tested at multiple graphics settings to allow testing more games. However, I could see the value in testing one different type of game (i.e. either Assassin's Creed or Tomb Raider) to provide that extra information for those who don't play FPS's.

  19. Hi mate i have same card and i have question i was using it with that support bracket but today i took it off i think it looks much better without it and i dont see any sag at all is that good or should i put it back for long time use? Thx for answer ;D

  20. folks just get the cheapest all models of the same gpu is always the same its a gimmick I got the cheapest 2070 just like the 1080 they both the same card but its always been that way diff brands its just names and looks the performance is always the same dont keep falling for the gimmick just get the cheapest

  21. So, should I buy MSI RTX 2700 gaming, not x and not z and OC it manually to save some bucks, or maybe even invest just in RTX 2700 Armor and OC it?

  22. Just ordered this card. Have to wait till Tuesday now. So, I am just watching more videos about it. lol

  23. i get on max settings with my 2060 gaming z 6gb playing fortnite on 1080p 137-139 fps so why do you get less? even with 1440p i get 112fps or a bit less so i dont understand this and then this: shows us the stats of fortnite while on the background playing fortnite and only gets 72fps XD this video is a joke

  24. In the uk i can get a gtx 2070 for cheaper than a 1080. Should i go for 2070 if i won't use ray tracing much if at all?

  25. Just spotted this card on ebay for AU$695 (May 2019). I have a GTX 970 now so this is tempting. But I prefer Gigabyte brand… I like the Windforce design, so need to think about it some more! This card looks nice though… decisions decisions.

  26. Hey, i know this is quite an old Video, but i just want to know if a 500W Power Supply is still good for this rtx 2070 or if I should upgrade to 550W or 600W. Im really not sure because some say 500W is enough, others say its not.

    Other specs: Just a Ryzen 5 2600X and some fans. No AIO or anything too spectacular installed.

  27. Is it worth upgrading from a msi gtx 1070…been debating on ordering it for the past few days. Also I dont plan on doing SLI

  28. Thinking to upgrade to this one from a gtx970. Currently I'm playing on a AOC 1440p,144hz monitor and considering switching to an Alienware 1080p, 240hz. Would that be a good idea having in mind that my GPU would be the msi 2070 gaming Z?!
    Also (my specs):
    I7 6700k (stock)
    Vengeance lpx 3.000Mhz 32gb
    280 SSD
    Asus maximus VIII hero

  29. Would a 600 watt power supply be able to handle this and a intel core i7-8700k with a corsair h100i pro? or should I go with like a 700 watt?

  30. Hey Jarrod I want to buy this card but on amazon it’s 560 pound and I found a cheaper one below it on amazon and it says msi 2070 gaming z boost graphics card is that the same one or is it a scam

Leave a Reply

Your email address will not be published. Required fields are marked *

Copyright © 2019 Explore Mellieha. All rights reserved.