Lack of testing makes it hard. Zen 5 is throughput monster so can see it being better late game Stellaris for example. But mid game while the dataset fits in V-Cache I could see the Zen 4 being faster.
So the X3D parts are going to be a mid cycle refresh before Zen 6. I mean if there is no competition then there is no reason not to slow play it and get people to double dip but it does not make me happy.
The cynic in me wonders if this will widen the gap between the 9700X and 9950X in games and that will push some gamers to go for the 9950X before the x3D parts launch in 6 months or so.
Not really.
The main thing is that if I buy a 6xx mobo now the board won't support Zen 5 and I need to update the bios. Shouldn't be an issue with usb flashing even without a CPU but still a pain for a new builder.
Having 800 series means OEMs get something new to sell, can iterate on designs...
Given you have said how CUDA proliferation is a problem for ROCm are you really unable to see the vast amounts of software written for x86 that has not been compiled for other architectures, really?
x86 gives them a different slice of the market and a very big slice, one they cannot compete in...
x86. Would get them more design wins for their CPU + GPU products like GH. Also gets them fabs and with how ruthless JHH is I bet he could turn Intel around pretty damn fast.
Hobbyists who want to play with a lot of MT performance but they don't actually make any money from it so don't want to spend a lot. IE a super tiny niche of the market.
Adroc and Kepler have specifically mentioned spec int 1t 2017.
Others are probably talking generally so will be going by the geomeaned average. Thing is if Zen 3 and Zen 4 are anything to go by then the geomeaned average is just a range of results that fit the SpecINT 2017 NT result so they are...
Nah.
Left: We want to get everyones basic needs met but cannot agree on which needs to do 1st so will splinter into our own niches and divide our power.
Right. We want to exterminate entire social groups but cannot agree on which order to exterminate them in, oh well, lets band together...
No. You just can't see past your view point.
A 1070 was good value. It has had a great life span and the MSRP was balanced with it's performance level. The 3060Ti less so.
New 8GB cards are value limited because they have performance anomalies. Those anomalies will increase in frequency as...
This is the point of the thread.
8GB parts have limited value due to compromises users need to make with them. Same applies to lower tier cards due to compute limits. The 3060 12GB MSRP was fine for its performance level so the price/perf/lifespan expectations were in line.
The...
I may be utterly wrong but I presume Keplers core for core comment was referring to a 96c Turin vs a 96c Genoa, would not really be core for core if core counts were not equalised.
Then it just means the inverse is probably true. Turin roughly matches Genoa in clock speeds but uses up the extra TDP budget to do that due to the IPC uplift. The 9950X having the same TDP as 7950X means it clocks lower in all core workloads. In ST workloads you don't max out the TDP budget so...
32% IPC in Spec Int 2017 1T. A very specific claim and it annoys me when people take that specific claim and try and generalise to all workloads, it screams dishonest and disingenuous.
Kepler Said
But the caveat is that Turing has 500W TDP Vs Genoa with 400W so with high core counts in an all...
Clockspeed, yields, node? 8c could be on an older node than 16c and 32c so would be quite a bit cheaper. V-cache compatibility.
Those are just off of the top of my head.
Still a compromise that the owner of the cheaper GPU does not need to make and that is the problem.
If they were good enough for their tiers they would not require the user to make more compromises than lower tier offerings. Yet in some scenarios they do so clearly they have an unnecessary...
The claim from adroc has always been 1t spec int rate 2017.
This test looks to have been an all core blender test.
They are not the same.
Only when we have 1t spec int rate 2017 results will we be know if the claim was accurate or not.
Businesses won't switch to nucs. Laptops are great all in one devices that make hybrid working a lot easier. Nucs don't solve anything at all and create a host of problems.
People will turn settings down obviously. The issue is that in case like TLOU at 1440p a 3060 12GB owner will need to make fewer compromises to get 'playable' fps than the 3060Ti 8GB owner will need to make or in something like Halo Infinite the 3060 12GB won't suffer from missing textures where...
Reviewers are part of the marketing wing, if they did not provide a net positive return the vendors would not spend money on it full stop. They do so clearly vendors get a benefit.
When it comes to economics if vendors cannot offer an attractive product at certain price points they are better...
AMD / NV / Intel and OEMs would not engage with reviewers if there was no benefit to their bottom line, it would be a waste of money.
$250+ is not ultra budget.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.