AMD Naples coming Q2 2017

[TheRandomness]

I like how there needs to be 6 notches to hold the CPU in place in the socket. I vote for 1974 pins in the socket (give or take 100)

[cj09beira]

2 minutes ago, MakeAMDGreatAgain said:

 

Love how they show benchmarks where the AMD CPU has like 20 more cores... Oh and it performs better!  

the thing is you cant buy a  2s server with more cores from intel, this is the best intel can offer in two cpus

[Fetzie]

54 minutes ago, MakeAMDGreatAgain said:

 

Love how they show benchmarks where the AMD CPU has like 20 more cores... Oh and it performs better!  

Did you love the two benchmark runs before then, when they had both at 44 cores and the AMD cpus whipped the 2699v4's asses (even when the memory on the Naples machine was running 2 JEDEC speed levels lower than stock)?

[JoeCoke]

7 hours ago, DocSwag said:

I know right, look at that Sandy Xeon.

Obviously, if you couldn't tell it was a sandy Xeon, then you don't know anything about computers.

[Agosto]

AMD with Naples ---> "Ce ripigliamm' tutt' chell' che è 'o nuostr'"

[MakeAMDGreatAgain]

22 minutes ago, Fetzie said:

Did you love the two benchmark runs before then, when they had both at 44 cores and the AMD cpus whipped the 2699v4's asses (even when the memory on the Naples machine was running 2 JEDEC speed levels lower than stock)?

Yeah, those were decent. But no point in doing benchmarks when it's not a fair comparison imo.

[Fetzie]

1 minute ago, MakeAMDGreatAgain said:

Yeah, those were decent. But no point in doing benchmarks when it's not a fair comparison imo.

Same core count, same memory speed seems pretty fair IMO. At the end it was just "look at what we can do that Intel can't".

[JoeCoke]

2 minutes ago, MakeAMDGreatAgain said:

Yeah, those were decent. But no point in doing benchmarks when it's not a fair comparison imo.

It's completely fair, as there is no other 32 core CPUs. Using those extra cores is demonstrating that and the advantages that come with it.

[MakeAMDGreatAgain]

24 minutes ago, Fetzie said:

Same core count, same memory speed seems pretty fair IMO. At the end it was just "look at what we can do that Intel can't".

No it's not the same core count though. The benchmark was 44 vs 64.. That's NOT a fair comparison, regardless of price.

 

I already admit, in the same core count the Intel version got smashed, but come on, don't distort benchmarks just because it's "AMD". Or think it's okay. There needs to be a baseline and level playing field.

[Fetzie]

Just now, MakeAMDGreatAgain said:

No it's not the same core count though. The benchmark was 44 vs 64..

 

44 cores vs 44 cores, 1866 memory on both sides.

[Spork829]

I've said since Ryzen was announced that they should release a Ryzen R9 line that comes after Xeons if what they're going for is tons of cores. Glad it's happening.

[iRileyx]

7 minutes ago, MakeAMDGreatAgain said:

but come on, don't distort benchmarks just because it's "AMD". Or think it's okay. There needs to be a baseline and level playing field.

If this was the other way around and amd only had 44 cores, where intel had 64, people would probably still want to see the 64v44 benchmark, when it would more than likely obviously be the 64 core that would win, its being curious and wanting to see just how good it can perform.

 

[MakeAMDGreatAgain]

15 minutes ago, Fetzie said:

 

44 cores vs 44 cores, 1866 memory on both sides.

Yep, I understand that one. But the one right after showed the 44 vs 64. No reason to as it's an unfair comparison.

[Fetzie]

2 minutes ago, MakeAMDGreatAgain said:

Yep, I understand that one. But the one right after showed the 44 vs 64. No reason too as it's an unfair comparison.

The first was a "this is a fair comparison because we're limiting ourselves to the Intel core count and memory speed". The second one was "this is what happens when we don't limit ourselves to the Intel CPU spec" (i.e. the "this is why you should buy Naples CPU" pitch).

 

Or shouldn't the 1080ti be compared to a 1080, because it's not a fair comparison?

[MakeAMDGreatAgain]

2 minutes ago, Fetzie said:

The first was a "this is a fair comparison because we're limiting ourselves to the Intel core count and memory speed". The second one was "this is what happens when we don't limit ourselves to the Intel CPU spec".

 

Or shouldn't the 1080ti be compared to a 1080, because it's not a fair comparison?

Red herring, not relevant unfortunately. Bottom line is 44 vs 64 cores is unfair

[Fetzie]

3 minutes ago, MakeAMDGreatAgain said:

Red herring, not relevant unfortunately. Bottom line is 44 vs 64 cores is unfair

So the R7 1800X shouldn't have been benched against the 6950X? Heck, by your logic it shouldn't have been benchmarked against the 400MHz faster 7700K. Because that would be "unfair". Boo fucking hoo. That's just how comparing products works.

 

Anyway. Let us see something representative. Some massive SQL data crunching, for example.

[Doobeedoo]

That looks epic though.

[MakeAMDGreatAgain]

7 minutes ago, Fetzie said:

So the R7 1800X shouldn't have been benched against the 6950X?

 

Anyway. Let us see something representative. Some massive SQL data crunching, for example.

Well, technically it's still an unfair comparison. 10 cores, 20 threads vs 8/16 threads. Even if the Ryzen performed better, it's still not an appropriate benchmark.

 

That's the problem that happens a lot these days in reviewers. The benchmarks become skewed bigly and don't use fair cores and threads comparatively. 

 

Look, I'm not an Intel shill or AMD one. As the first benchmark hey had the same cores, but then AMD decided to do the 44 for 64 benchmark which wasn't fair 

[Fetzie]

Just now, MakeAMDGreatAgain said:

Well, technically it's still an unfair comparison. 10 cores, 20 threads vs 8/16 threads. Even if the Ryzen performed better, it's still not an appropriate benchmark.

 

That's the problem that happens a lot these days in reviewers. The benchmarks become skewed bigly and don't use fair cores and threads comparatively. 

 

Look, I'm not an Intel shill or AMD one. As the first benchmark hey had the same cores, but then AMD decided to do the 44 for 64 benchmark which wasn't fair, or appropriate. 

Why isn't it appropriate when they had to disable a quarter of their cores and reduce the memory speed from 2400 MHz to 1866 MHz to get it to be "fair" in your eyes? Why can't AMD show what their product can do when compared to the best dual-socket CPU that Intel will sell you?

[MakeAMDGreatAgain]

Just now, Fetzie said:

Why isn't it appropriate when they had to disable a quarter of their cores and reduce the memory speed from 2400 MHz to 1866 MHz to get it to be "fair" in your eyes? Why can't AMD show what their product can do when compared to the best dual-socket CPU that Intel will sell you?

Well, you have a good point tbh. I'm just saying it looks bad on AMD to show benchmarks like "44 vs 64". When they are already smashing the competition with equal amount of cores. It just doesn't feel right to do that (or why do they have to)? People will just use that benchmark now to complain that AMD is doing "unfair benchmarks". When in reality, they squashed them at equal cores, so why show the unfair one? Just to give AMD shills more ammo?

 

 

[MoonSpot]

14 minutes ago, MakeAMDGreatAgain said:

Red herring, not relevant unfortunately. Bottom line is 44 vs 64 cores is unfair

Please stop talking.

[MakeAMDGreatAgain]

Just now, MoonSpot said:

Please stop talking.

Okay, please tell me why 44 vs 64 is a fair comparison benchmark, when AMD already beat them on equal amount of cores?

 

Also, btw it's "posting", last time I checked, I cannot talk into a textarea

[WMGroomAK]

2 minutes ago, MakeAMDGreatAgain said:

Okay, please tell me why 44 vs 64 is a fair comparison benchmark, when AMD already beat them on equal amount of cores?

 

Also, btw it's "posting", last time I checked, I cannot talk into a textarea

Problem is, the 2699 v4 is the only direct competition on a core to core count for Intel until they release they 2699 v5. I don't see why AMD should not show both the Intel and AMD CPUs operating at the most cores they have.

[MakeAMDGreatAgain]

3 minutes ago, WMGroomAK said:

Problem is, the 2699 v4 is the only direct competition on a core to core count for Intel until they release they 2699 v5. I don't see why AMD should not show both the Intel and AMD CPUs operating at the most cores they have.

That's true, but just to me personally, it gives off a vibe of unprofessionalism when showing unfair benchmarks. When in reality, their recent benchmark right before that demolished it (and was a fair comparison). Doesn't make sense to me, but whatever.

 

They are just giving more ammo to Intel shills, doing bologna shit like that imo.

[Fetzie]

8 minutes ago, MakeAMDGreatAgain said:

Well, you have a good point tbh. I'm just saying it looks bad on AMD to show benchmarks like "44 vs 64". When they are already smashing the competition with equal amount of cores. It just doesn't feel right to do that (or why do they have to)? People will just use that benchmark now to complain that AMD is doing "unfair benchmarks". When in reality, they squashed them at equal cores, so why show the unfair one? Just to give AMD shills more ammo?

 

 

Enterprise (as in the people that will actually buy these processors) doesn't listen to the people that might bitch about how a benchmark is unfair.