(16core added)AMD 3000 specs! 4.7 GHZ, R9 3950x, R7 3700x, 3800x.

[RejZoR]

I love people who go absolutely crazy about NVIDIA's superiority in performance and then go on and buy a fucking GTX 1650Ti to really celebrate the hype they are doing on forums... Mid end segment is where most people buy cards and that's where AMD is doing just fine. Bottom end are only bought by clueless people or for systems that don't have GPU on the CPU itself (like really high end workstations that only need CPU for workloads).

[Xkillerpn]

16 minutes ago, RejZoR said:

I love people who go absolutely crazy about NVIDIA's superiority in performance and then go on and buy a fucking GTX 1650Ti to really celebrate the hype they are doing on forums... Mid end segment is where most people buy cards and that's where AMD is doing just fine. Bottom end are only bought by clueless people or for systems that don't have GPU on the CPU itself (like really high end workstations that only need CPU for workloads).

ATM as is stands at least in the uk a vega 56 is around 240 to 270 the nice ones, to get the same performance on nvidia it will be 300+. and lets not forget thats a 6g cards, vegas have 8g and HBCC. 

 

The 580 is really cheap atm, on 1080p or 1440p amd has the uper hand on budget.

 

Now i dont care about power but for 6 hours of play a day or 4 its not going to make a diference. 

[Quadriplegic]

13 hours ago, Senzelian said:

I'm happy that they did, but I wish they beat them by a larger margin.
1% better single core performance and 2% better multi-core performance compared to a now almost 1-year-old CPU, which is based on an aging architecture just isn't interesting, even tho this is AMD's own chart!

 

 

That's why I said, at least the price is right.
 

I think PCI-e 4 and the security issues on Intel chips are a much more compelling reason to buy a 3rd gen Ryzen CPU. 

Keep in mind 9700k boosts to 4.9 Ghz, 3700x boosts to 4.4 Ghz. 

The future of Zen 2 chips rely solely on their overclocking headroom, if they can't overclock past 4.5-4.6, Intel will remain superior. 4.8 Ghz or higher on Zen 2 will give AMD lead for a while, considering Intel can't do much with their current architecture.

[Trixanity]

15 minutes ago, Quadriplegic said:

Keep in mind 9700k boosts to 4.9 Ghz, 3700x boosts to 4.4 Ghz. 

The future of Zen 2 chips rely solely on their overclocking headroom, if they can't overclock past 4.5-4.6, Intel will remain superior. 4.8 Ghz or higher on Zen 2 will give AMD lead for a while, considering Intel can't do much with their current architecture.

Unless Intel intends to backport their stuff it could be a while. Intel's roadmaps don't show anything but 14nm over the next 2-3 years. They have a lot of 14nm for desktop in the pipeline but they're really limited in what they can do.

[MadDuke]

Everything (gaming wise) relies solely on the Overclocking headroom.

 

These chips were NEVER expected to run freakishly high clock out of the box because you still need to be able to slot even the 16 Core 32 Thread beast into an existing B350 motherboard and not have the VRM blow in your face.  (talking stock, if you overclock on most of them it will happen). 

[Senzelian]

4 hours ago, Quadriplegic said:

Keep in mind 9700k boosts to 4.9 Ghz, 3700x boosts to 4.4 Ghz. 

The future of Zen 2 chips rely solely on their overclocking headroom, if they can't overclock past 4.5-4.6, Intel will remain superior. 4.8 Ghz or higher on Zen 2 will give AMD lead for a while, considering Intel can't do much with their current architecture.

The average consumer doesn't overclock.

[rockking1379]

18 hours ago, kiwibacon said:

3800x is going to be the best gaming chip in the line up. Clearly 3900x is cut down 16 core. While 3800x is higher binned 3700x

But is that worth the money over the 3700x will be the question. 

I personally don’t think it is. But then again not everyone is comfortable messing in bios to get an overclock done. 

[ThePD]

5 minutes ago, rockking1379 said:

I personally don’t think it is. But then again not everyone is comfortable messing in bios to get an overclock done. 

Totally correct, there are many people who buy 2080 graphics card in products and don't even know that a 2080 is a Nvidia product. I'd say a vast majority of people do not know how to overclock or even know what overclocking is.

[MadDuke]

8 minutes ago, ThePD said:

Totally correct, there are many people who buy 2080 graphics card in products and don't even know that a 2080 is a Nvidia product. I'd say a vast majority of people do not know how to overclock or even know what overclocking is.

I wholeheartedly agree. But then agian. Those people don't care about +-10% performance and would be pleased with an 2700x strictly for gaming even. 

[RejZoR]

56 minutes ago, MadDuke said:

Everything (gaming wise) relies solely on the Overclocking headroom.

 

These chips were NEVER expected to run freakishly high clock out of the box because you still need to be able to slot even the 16 Core 32 Thread beast into an existing B350 motherboard and not have the VRM blow in your face.  (talking stock, if you overclock on most of them it will happen). 

You don't need to run it at 5GHz if 4.6GHz gives same results thanks to IPC increase. It seems AthlonXP history is repeating where clock isn't everything. Intel was boosting clocks to 3.2GHz back in the day and AthlonXP processors had only 2333MHz (Athlon XP 3200+, Barton core) clock and could easily rival 3200MHz Pentium 4's. Yet people were again only looking at the clock. AMD made the + scale to somewhat counter that. And I can tell you it wasn't a lie. I had the AXP 2400+ that put 2400MHz Pentiums to shame. At only 2000MHz! And with overclocking I got it up to 3200+ performance.

 

So, if Ryzen 9 3900X 4.6GHz delivers same performance as Intel's CPU at 5GHz, then this is exactly that. 4.6GHz to 5GHz is around 8% clock difference (I doubt it's also 8% performance increase). AMD is claiming up to 15% IPC boost over Zen+. And knowing Zen+ (Ryzen 7 2700X) was already on par with Intel in IPC department, AMD actually being faster than Intel at lower clock isn't that suprising because that's what IPC is. More work done at same or lower clock. Or in worst case scenario, is at least the same.

 

I too was hoping for 5GHz Ryzens, but seeing how they make up the clock difference with IPC, it's looking really interesting.

[Stefan Payne]

6 hours ago, Quadriplegic said:

Keep in mind 9700k boosts to 4.9 Ghz, 3700x boosts to 4.4 Ghz. 

We don't know how far it boosts.

I've seen my 1700X boost to 3,9GHz for a short while.

That's the official boost. We need to wait and see how it will be in Reality, if there is a short term boost beyond that...

 

 

[Stefan Payne]

6 hours ago, laminutederire said:

It's not good business to not move on when needed. No one expects them to beat nvidia anymore anyway, and certainly not while still on gcn like architectures, so it would be wise to  focus on them.

GCN (Shader) Architecture is fine. That is said by a ton of developers (source: German Forum 3DCenter.de)

If GCN was so bad, nVidia wouldn't have adoptet more and more of it with each generation. 

 

The Problem AMD has is the frontend, triangle throughput and some people say Compression or other memory saving things. Though it might also have something to do with the drivers - and why nVidia doesn't allow open Source Drivers for their chips. Because then you could compare them on equal terms. Something nVidia hates to do, wich is why they stopped supporting OpenCL once GCN was released.

[Trixanity]

It's possible that XFR and/or X570 boards can clock them beyond their rated spec either manually or automatically. It's also possible the frequencies we see are based on an otherwise steep voltage curve after 4.6 GHz to avoid another Vega scenario with chips clocked beyond the sweet spot. It's also possible that it simply can't clock above 4.6 GHz (or perhaps a 4.7 GHz with a big voltage increase) at all. We'll see. Anything's possible. Bench for waitmarks.

[Stefan Payne]

11 hours ago, Humbug said:

Not for the general public but for particular groups. By having more vulnerabilities exposed continuously over the last couple of years it has eroded trust in Intel as a secure platform and their engineering capabilities, or raised suspicions about the shortcuts they make for performance.

It is just the start of it...

And gives fodder against Intel.

It shows that Intel is not as good as people claimed and IIRC someone somwhere said (was it Barnaclys??) that Intel reduced their Test Procedure to be able to push out more products in a shorter period of time to be "more competitive". That was after the FDIV Bug where Intel even had to replace the affected processors.

 

 

But now people can not claim that Intel is "more stable" or in general "better" than others are. That will be a problem in the future, if AMD can keep up their development of Zen with Zen2+ or 3...

 

Back with K7 AMD was developing it further into the K8 but after that their development stopped and all they brought was the mediocre Phenom. Back in those days AMD's Problem was the Cache. It was slow as hell. Especially the Bulldozer was affected badly by an awfully slow cache that crippled its performance....

+it needed Software optimized for the Architecture because the Architecture was too different from Intel...

 

Though the irony is that the one to put the memory controller into the CPU threw it out first X-D

 

 

And I know that back in the day, People still bought Intel despite beeing far worse because it had SMT and they claimed that it "felt more fluidly". and other stuff.

 

 

If AMD has lower power consumption (just like in the K8 Era), higher performance, what reason is there to still buy Intel??

The "Intel is more stable" thing is just nonsense, especially in the Desktop Enviroment. And the effects it will have are not forseable...

[RejZoR]

14 minutes ago, Trixanity said:

It's possible that XFR and/or X570 boards can clock them beyond their rated spec either manually or automatically. It's also possible the frequencies we see are based on an otherwise steep voltage curve after 4.6 GHz to avoid another Vega scenario with chips clocked beyond the sweet spot. It's also possible that it simply can't clock above 4.6 GHz (or perhaps a 4.7 GHz with a big voltage increase) at all. We'll see. Anything's possible. Bench for waitmarks.

I think 4.6GHz is a hard locked boost. At least there wasn't anything mentioned above going higher than that. Unless they kept it under the radar on purpose and reviews will uncover that they can do that if conditions are acceptable for it (like how NVIDIA does where they state the guaranteed boost, but they always clock far beyond that out of the box). But I don't think that's the case here, otherwise they'd brag about it during the keynote how boost is more intelligent and it can do that. But they haven't mentioned a thing.

[Trixanity]

11 minutes ago, RejZoR said:

I think 4.6GHz is a hard locked boost. At least there wasn't anything mentioned above going higher than that. Unless they kept it under the radar on purpose and reviews will uncover that they can do that if conditions are acceptable for it (like how NVIDIA does where they state the guaranteed boost, but they always clock far beyond that out of the box). But I don't think that's the case here, otherwise they'd brag about it during the keynote how boost is more intelligent and it can do that. But they haven't mentioned a thing.

We still lack a lot of details. It seems like AMD is forcefully splitting the info across two events (probably for the hype factor) so we might hear more on the 10th (if not, then 7/7). They haven't mentioned XFR yet so either it's deprecated or it can't do anything for Zen 2. 

So if there is any more frequency left on the table I actually think it's smart to wait until the 10th to unveil that. Will be all the more of a spectacle that way. Although it's not like I'm expecting anything just so we're clear. I fully expect it to top out at 4.6 until I hear anything else. 

[thechinchinsong]

3 hours ago, MadDuke said:

Everything (gaming wise) relies solely on the Overclocking headroom.

 

These chips were NEVER expected to run freakishly high clock out of the box because you still need to be able to slot even the 16 Core 32 Thread beast into an existing B350 motherboard and not have the VRM blow in your face.  (talking stock, if you overclock on most of them it will happen). 

Gaming wise definitely doesn't completely rely on overclocking headroom. You could overclock FX series chips 20 to 25% (up to 5-ish ghz) and still have terrible gaming performance. I know this is an extreme example, but I'm just pointing out that "IPC" usually matters as much as clock speeds.

[thorhammerz]

3 minutes ago, thechinchinsong said:

 I know this is an extreme example, but I'm just pointing out that "IPC" usually matters as much as clock speeds.

Performance = IPC x Clock Speed.

 

a.k.a.

 

(Instructions per unit of time) = (Instructions per Cycle) x (Cycles per unit time).

 

Spoiler

Extremely simplified, assuming no bottlenecks from I/O or elsewhere.

Of course, clock speed makes much better marketing material, so by and large it'll be the number touted to the masses.

[Mihle]

1 hour ago, Trixanity said:

We still lack a lot of details. It seems like AMD is forcefully splitting the info across two events (probably for the hype factor) so we might hear more on the 10th (if not, then 7/7). They haven't mentioned XFR yet so either it's deprecated or it can't do anything for Zen 2. 

So if there is any more frequency left on the table I actually think it's smart to wait until the 10th to unveil that. Will be all the more of a spectacle that way. Although it's not like I'm expecting anything just so we're clear. I fully expect it to top out at 4.6 until I hear anything else. 

I bet E3 will be more focused on Navi than Ryzen.

[The Benjamins]

6 minutes ago, Mihle said:

I bet E3 will be more focused on Navi than Ryzen.

Computex 80% ryzen 20% navi, E3 80% Navi, 20% Ryzen

[Trixanity]

53 minutes ago, Mihle said:

I bet E3 will be more focused on Navi than Ryzen.

Definitely. Anything else would be crazy given the audience.

[mr moose]

12 hours ago, laminutederire said:

It's not good business to not move on when needed. No one expects them to beat nvidia anymore anyway, and certainly not while still on gcn like architectures, so it would be wise to  focus on them.

It is very bad for a company to rush out mediocre products at any time.   Even Intel and Nvidia when they had no competition from AMD (for wither GPU or CPU) never pushed out mediocre, it was always close to the best they could produce.   It's just not good business practice to do it regardless if you expect them to be competitive or not.

 

11 hours ago, RejZoR said:

I love people who go absolutely crazy about NVIDIA's superiority in performance and then go on and buy a fucking GTX 1650Ti to really celebrate the hype they are doing on forums... Mid end segment is where most people buy cards and that's where AMD is doing just fine. Bottom end are only bought by clueless people or for systems that don't have GPU on the CPU itself (like really high end workstations that only need CPU for workloads).

 

Don't forget that just because someone can only afford stupid pointless cards doesn't mean they don't understand the industry and are still interested in discussing why NVIDIA are at the top.   I have a 570,  I bought it in full knowledge that it is basically at the bottom of the mid range,  but that does not mean I don't care about the top end or what it means for business who strive to be there. 

[Taf the Ghost]

4 hours ago, RejZoR said:

I think 4.6GHz is a hard locked boost. At least there wasn't anything mentioned above going higher than that. Unless they kept it under the radar on purpose and reviews will uncover that they can do that if conditions are acceptable for it (like how NVIDIA does where they state the guaranteed boost, but they always clock far beyond that out of the box). But I don't think that's the case here, otherwise they'd brag about it during the keynote how boost is more intelligent and it can do that. But they haven't mentioned a thing.

 

4 hours ago, Trixanity said:

We still lack a lot of details. It seems like AMD is forcefully splitting the info across two events (probably for the hype factor) so we might hear more on the 10th (if not, then 7/7). They haven't mentioned XFR yet so either it's deprecated or it can't do anything for Zen 2. 

So if there is any more frequency left on the table I actually think it's smart to wait until the 10th to unveil that. Will be all the more of a spectacle that way. Although it's not like I'm expecting anything just so we're clear. I fully expect it to top out at 4.6 until I hear anything else. 

It looks like Hot Chips in August is when we finally get the deep technical details beyond things like Boost Tables which should happen at launch.

[Stefan Payne]

Since people claimed that MSI won't support Zen2...

 

Meanwhile, MSI:
https://www.msi.com/Motherboard/support/X370-GAMING-PRO-CARBON.html

 

7A32v1N3(Beta version)

Description

- Update AMD ComboPI1.0.0.1
- Support new upcoming AMD cpu.
Note: 
Do not update the BIOS when using the cpu listed below.
A8-9600/A6-9500E/A6-9500/A12-9800E/A12-9800/A10-9700E/A10-9700/970

 

ComboPI1.0.0.1 was seen with other Manufacturers showing PCIe Gen 4 in the BIOS...

 

Others also come with that BIOS Version, such as B350 Gaming PRO CARBON
https://www.msi.com/Motherboard/support/B350-GAMING-PRO-CARBON.html

 

B450 KRAIT:
https://www.msi.com/Motherboard/support/B350-KRAIT-GAMING.html

[leadeater]

10 minutes ago, Stefan Payne said:

Since people claimed that MSI won't support Zen2...

One of the videos I watched mentioned that there are MSI motherboards that won't support Zen2, think that list is rather small though. Don't remember who's video it was.