Gadgetory

this is a 1070 TI duke card from msi this card is supposed to be priced at 472 480 dollars and it was when we originally filmed this review but we had to add this part in because now it's $900 which makes it absolutely not at all worth it however in the event the prices come back down to 472 480 this review should serve as your guide so if you're watching this just basically use the review if the prices are reasonable and if the prices are still 800 + then don't buy the card at all period and this one lands just between 1080 s and the baseline 1070 ti reference card with of course vega 56 nebulously in the area depending on how retailers feel at that day before that this video is brought to you by thermal grizzly makers of the conductor not liquid metal that we recently used to drop 20 degrees off of our coffee leak temperatures thermal grizzly also makes traditional thermal compounds we use on top of the IHS like cryo not and hydro not pastes learn more at the link below a couple of things here about this video card this one might be appealing to you because it's a three fan cooler and it's reasonably priced the cooler however is running a 2 slot form factor configuration so a lot of the larger fan coolers on the market lately have been two and a half or even slightly more for the expansion slot size and that means that although it will fit in a lot more builds more easily it is using a much thinner radiator or heatsink if you prefer in fact it it's significantly thinner it's about half the thickness of what we saw on the recent 2.5 it slot design for the EVGA 1070 TI ultra silent or whatever they're calling it so quite a bit different MSI has tended to stick to two slot designs there Twin Frozr coolers typically have used two slot designs although there are some that have gone larger and this one is what they're calling a tri froze our cooler so three fans dry frozer it is not really as simple as three fans equals better though it's not more more is more better because the Twin Frozr fans are taller and a bit wider so this is just going for three smaller fans instead as for the advocacy of the cooler we're really focusing on that today with video cards as many of you know at this point if you've seen our channel the difference between tarda and card be within the same class ie 1070 TI is pretty small in fact the difference in gaming performance out of box is clock for clock far more likely to vary based on the silicon quality than it is based on the actual card or the vendor itself if msi makes one of these an EVGA makes a similar one the difference between them gaming wise and FPS wise is almost certainly going to hinge more on which one had the better silicon when you bought it so for that reason we're focusing on the thermal cooling ability is the noise levels and some overclocking potential versus the colorful 1070 TI volcán x that we reviewed all of which will come together to help us understand how good of a card it actually is we also have vrm information coming out hopefully somewhat soon that will be done by builds Oya if you want to see the power component analysis from an overclockers perspective let's get into thermals first for the full testing methodology as always check the article linked in the description below that contains our test bench and our test methods starting with a complete stock auto test for out-of-the-box thermals we measured the gtx 1070 TI duke at about 49 degrees celsius over ambient and again that's a delta t number so it is del z arraignment for the GPU diode rating we're at 49 and this was during our standardized power buyers torture test the MOSFETs operated at about 51 degrees Celsius over ambient with the hottest gddr5 module measuring at 65 degrees for the MOSFET and memory measurements we are using k-type thermocouple that are 1/100 of an inch thick they are laser thin for a direct component contact measurement and this is effectively a case measurement we don't have many 1070 TI cards tested yet but these measurements place the MSI Duke markedly behind the only other 1070 Ti currently on our bench the colorful Vulcan X the Vogon X operated about 8 degrees cooler on the GPU temperature 2 to 3 degrees cooler for the MOSFETs though those aren't direct comparison since there are different vets and about three to four degrees warmer for the memory keep in mind that MOSFET temperature here is largely irrelevant because all we're really going for is making sure it's within spec which these are they are well under spec in fact for both of them although one is more under spectrum the other the Vogon acts had an unimpressive base plate and vram cooling solution so these numbers make sense where MSI is a bit better for the vram cooling the Duke is one of the warmest cards on the bench though and part of the reasoning is because Emma size fan profile and V bios simply isn't aggressive enough and the fans are kind of weak to begin with during Auto testing the fans tend to stick to around 38 to 42 percent speed with an ambient temperature of about 21 to 22 Celsius and this has them at about 30 decibels of noise that's way too conservative we've noticed that MSI seems to target about 66 degrees Celsius for the core temperature and so they are adjusting fan speeds based on that target this means that memory can sometimes run warmer because the fans don't need to spin very fast for the GPU to be 66 degrees but the vram doesn't have any governance over the fan speeds here are some noise normalized numbers with thermals tested at 40 decibels normalized for all tested cards the eye-catching thing here is that thermal performance improved the one set to 40 decibels output which is untrue for nearly every other card on the bench this happens only for one reason the fans are spinning faster at 40 decibels despite 40 decibels being generally on the lower end of video card noise emissions the Duke performs about 8 degrees warmer than the Vulcan acts when both are at 40 decibels with a GPU temperature of 45 point 4 degrees Celsius delta T over ambient we're also seeing a mosfet temperature of 45 degrees on the MSI Duke Card about the same as what the Vulcan X saw and a 58 degree memory temperature whereas colorfull ran 64 degrees speaking of noise here's a look at that noise levels show the msi 1070 TI duke toward the very bottom edge of the pack when at each rpm level and this is a PWM signal to noise response curve most graphics cards like to keep their fans at around 50 to 55 percent so we'll highlight that area of the decibel output and that is used as a hard target in B BIOS general the msi cards fans don't spin very fast and tend to max out at around 3300 RPM they're admittedly very quiet for the most part until you hit a hundred percent or so but they are also ineffective at cooling again considering 40 decibels was a faster speed than auto the tuning is much different from most cards on the market as for overclocking here's our overclocked stepping chart we found the msi duke card to be stable briefly at 225 megahertz core and 500 megahertz memory we ultimately began crashing in games with those frequencies though so we had to step down to 200 megahertz offset core and 450 megahertz offset for the memory again these are offsets not the actual speeds the clock averaged about 20 50 megahertz in this configuration which is one part power limit and one part thermal limit Pascal increases clock speed for roughly every 5 degrees Celsius on the core and if the Duke were capable of cooling itself below 60 degrees in our room ambient of about 20 22 C we'd see a couple megahertz higher clock we also ended up peaking at around 20 88 megahertz though hit a 21-14 megahertz core speed for about half a second before crashing that was with a 250 megahertz offset which ultimately did not take here's our colorful vulcan XO as he table for reference for this one we got stuck at around 20 88 megahertz table average clock or 2120 six megahertz peak frequency the results is better for the Vulcan X and the resulting performance and fire strikes shown here with fire strike Ultra comes out to 53 66 points versus 53 15 points for graphics this was done on repeated testing and is outside of error as we ran the test numerous times an average did fire strike Xtreme has us about 200 points different as well once again favoring the colorful card as you can see here and now comes the usual reminder these clocks don't necessarily mean you can hit these clocks in fact you might do better you could do a lot worse the difference primarily will come down to the actual silicon quality much like the frequencies that you can hit out of box will largely come down to silicon quality not to the card itself or the design of the PCB or really anything else its silicon quality first then you start wearing things like cooling ability because you've got that thermal limiter to worry about and then from there power starts to come into play once you're overclocking but until then not so much so this card is an instance of a slightly below average cooler it's not awful they do run actually pretty quiet which is an upside relatively but there's a reason that EVGA when they did their ultra silent version fattened up the heatsink it's because if you're gonna run quieter fan speeds you do need something else to help deal with the heat it's generally more mass or more surface area more specifically speaking and this card if you kind of if we have b-roll shots of the side of it or anything you'll see just how thin that heatsink is it's comparatively really just about the same size as the heat pipes so it's a lot smaller than what you see normally and that's where the difference in thermal performance it's coming from before the most part the fans are a bit weak but they're not the worst because again they are at least somewhat quiet and MSI does as we show on our teardown contact some of the power components with at least they're all pads connected to the fins of the heatsink so you're getting some level of transfer some nominal level of transfer between the vrm components and the heatsink which is better than for example the zotac amp extreme that we looked at previously but it's just it's simply ok it's not that exciting and you know as a 10-7 ETI for the price it's not bad you can find better but it's also not the worst thing a messiah has ever made they we have we have examples of that on the channel if you're curious so yeah overall not too exciting or disappointing either way it is certainly warmer than the Vulcan X I'm not sure what the Vulcan X is selling out in the US has been in and out of stock we'll be looking at me 1070 TI cards from EVGA it's somewhat soon and hopefully one of them ASIS but other than that we've got some of these tricks and power colored Vega coverage coming up sometime hopefully before CES but we'll see the Intel bug is throwing us off a bit that video should already be online as well subscribe to the channel for everything as always you can find links to this and description below if you are interested in it and you can pick up one of our shirts on store that gamers nexus net or you can go to patreon.com/scishow and x's topside directly and of course the mod mat has been a big item for us as well store that gamers Nexus Don that mod Matt thank you for watching I'll see you all next time