Calculation force of 2080ti and 3060ti
is the "younger rival" of
rtx3070
, which is naturally slightly higher than
rtx3060ti
in terms of "brand", ASUS
is more "famous" than
Gengsheng
and its "after-sales guarantee" is more stable
PS: rtx3060ti
is ready to replace the "graphics card" of
rtx2080
Super
performance level.
but for me, whether it's the 30 series or the 20 series, my 10 series is still enough. After that day's end, I guess I'll change it. It's estimated that at that time, the 40 series and 50 series will come out. With the increase of age and pressure, I've missed the age of the best hardware, Now the most important thing is to have fun, but I still want to advise you to do what you can, whether it's graphics card or CPU, as long as you can meet your own requirements! What do you think of this 30 series graphics card? Welcome to comment area
Big
RTX 2080 uses a complete tu104-400 core, core basic frequency 1710MHz, non-public version 18000mhz, with 8GB 14ghz gddr6 video memory, the bandwidth reaches 448gb / s. In terms of TDP, the non-public standard version is 225W, while the public version is 215W
RTX 2070 uses a complete tu106-400 core, with a core base frequency of 1410mhz, an acceleration of 1683mhz for the public version and 1620mhz for the non-public version. With 8GB 14ghz gddr6 video memory, the bandwidth reaches 448gb / S. in terms of TDP, 175W for the non-public Standard Version and 185w for the public version
extended data:
precautions:
some CPUs need to be used with graphics cards, so pay attention to this when purchasing
the interface of the graphics card has the anti daze design. When installing the graphics card, pay attention to the position of the anti daze interface, and do not insert it wrong
the fan of the graphics card does not work, which may be that it does not meet the temperature standard
the graphics card is divided into public graphics card and non-public graphics card, which can be purchased according to the demand. The public graphics card will adopt the updated display chip in advance
the main reason is that there are more functional controls.
please adopt it.
major video websites (bilibilibili, Acfun, Youku, etc.) have detailed game actual screen evaluation of various graphics cards, which can be compared intuitively.
the above are the technical specification parameters that we can grasp at present. Some parameters can only be revealed through software testing after getting the real card, such as the number of transistors, texture units, ROPS. However, we can basically understand the performance improvement degree through the above parameters
through the parameters, we know that the number of stream processors of rtx2080ti is 4352, while the number of stream processors of the last generation flagship cahuang gtx1080ti is 3584, which is 20% higher than that of the last generation graphics card. We can conclude that the performance of rtx2080ti is 1.2 times higher than that of the last generation graphics card (at least). By analogy, the number of stream processors of rtx2080 is 15% higher than that of gtx1080, so the performance is improved by at least 15%
let's look at the type of video memory. Unexpectedly, it is the new gddr6. Although the capacity of video memory is consistent, considering that the bandwidth, frequency and power consumption of gddr6 are much better than gddr5x, the performance of gddr6 will continue to improve from this point of view, and the specific improvement compliance will be revealed by subsequent evaluation
IP accelerator www.ipdaili.com We know that the starting price of the gtx980 is $559, that of the gtx1080 of the previous generation is $599, and that of the rtx2080 of the new generation is $799, and that of the Bank of China is $6499. This is unacceptable to some players, although we know that the performance will definitely improve a lot, But the price has become more expensive. In this way, where is the cost performance advantage? Obviously, it is not as sincere as AMD
2080 can be understood as the addition of ray tracing and AI to
GTX
1080ti (of course, there are also differences in technology and architecture to make the performance slightly higher, and the performance of RTX
2080ti is much higher than that of
1080ti
1080ti
for rendering software, some software support home GPU acceleration (generally plug-in renderers). These graphics cards can be used, and the gap is not very big. All of them are home graphics cards, and the gap is not very big in double precision. Otherwise, rendering only needs to see
CPU's multi-core, multi line and high frequency
in addition, the feature of professional graphics card is driver, which needs software to support the acceleration of professional graphics card. In this way, the modeling speed will be much faster, and the modeling will not get stuck when building large-scale projects, otherwise it will be similar to home graphics card.
