AMD Ryzen 9 5900X

[Spanners]

The point is though, AMD may have raised core counts, but for mainly gaming, they are irrelevant. Imo games still do not need more than a 6/8 core cpu at most. Everyone raved when Ryzen came out, oo ooo but it has 12 cores(well 6 with hyper) but games didn't use them anyway. Imo the one advantage of ryzen coming out was the big kick up the behind it gave Intel. They may have been caught with their pants down, but do not write them off(as every Ryzen fan seems to be doing) they have the budget, and experience to recover from this, even though it may take them a while after been caught snoozing.

Sure I agree, beside some fringe cases 6/6 or 6/12 is still fine. In fact with a mid-range GPU that's not bleeding-edge something like a i5 4570 still means you're not missing out on much in the majority of titles (that you can play with a 2-4 year old mid-range GPU). There are some newer games however that use 8+ cores and you could make an argument that 8 cores is a smart move due to next gen consoles if you're planning on a 3-4 year upgrade cycle. That may turn out to be something that seems to makes sense but isn't actually the case but it's not crazy to infer.

This isn't just about gaming though. I personally don't game and care about DaVinci Resolve performance more than anything else at the moment and this seems to be a great CPU for me. Although compared to a 3950X it would not be much of an upgrade.

Not writing Intel off by any means whatsoever, I'll be very much looking forward to their response. Give AMD some huge credit here though. I would have laughed in your face if you'd told me this was the current state of the CPU market during the Bulldozer era.

 

[springs113]

"AMD, please save us from Intel and Nvidia's expensive pricetags."
Yeah. The same guys that were once screaming that line in previous years are now fine with AMD's new prices. Things sure as hell didn't get cheaper with AMD, and in fact they're matching the competition in prices and sometimes even more so: $1000 GPU, $4000 HEDT CPU, and now almost $1000 for a mainstream CPU. Hypocrisy at its finest.

So I guess a 6core intel part didn't cost $520? Or a 10 core part for $1730? My how we forget. A 1800x launched at $500 and that was a steal, these processors make that processor look like childs play yet you complain? SMDH

 

[agentnathan009]

"AMD, please save us from Intel and Nvidia's expensive pricetags."
Yeah. The same guys that were once screaming that line in previous years are now fine with AMD's new prices. Things sure as hell didn't get cheaper with AMD, and in fact they're matching the competition in prices and sometimes even more so: $1000 GPU, $4000 HEDT CPU, and now almost $1000 for a mainstream CPU. Hypocrisy at its finest.

So a possible $999 competitor to a $1,500 nVidia 3090 blows your argument out of the water. A 6 core CPU that gets within striking distance of Intel's highest end consumer part and is significantly cheaper blows you argument out of the water. Last, but not least, the computational power of that $4,000 HEDT CPU demolishes Intel's best 18 core HEDT part so badly that I cannot fathom why you would even complain about the price given how much less it costs and how much more performance you are getting. There is no hypocrisy, AMD is giving us more for less than Intel, and still giving us more for less.

I'm sorry that you can't buy a Ferrari for $10,000 brand new off the showroom floor because you think it shouldn't cost any more than that. How dare Ferrari charge $200,000 for a high performance sports car!

AMD, still providing better bang for you buck

 

[RMX]

I've said it before and I'll say it again, to all those wondering/arguing over TPU scores being lower vs. other reviewers: TPU tested with an older BIOS using AGESA 1.0.8.0, while others used the intended BIOS for Ryzen 5000, with AGESA 1.1.0.0 on their boards (AMD mentioned this in their videos).
Based on this, I tested it myself and it became clear there is a substantial difference between the two when running Cinebench R20 1T or games like SOTR.

 

[Fasola]

I've said it before and I'll say it again, to all those wondering/arguing over TPU scores being lower vs. other reviewers: TPU tested with an older BIOS using AGESA 1.0.8.0, while others used the intended BIOS for Ryzen 5000, with AGESA 1.1.0.0 on their boards (AMD mentioned this in their videos).
Based on this, I tested it myself and it became clear there is a substantial difference between the two when running Cinebench R20 1T or games like SOTR.

W1zzard said that it was an initial typo and that 1.1.0.0 was used in one of these Zen 3 review topics somewhere (reviews were updated as well). I expect a certain degree of copy paste is used in order to keep the review format.
At this point, we should just wait for the results of his investigation as all this pointless speculation is doing nobody any favours.

 

[Dave65]

Thank you for the review @W1zzard
yup... still not as fast as intel in games

Funny, most of the other tech tubers I watch has it ahead of the 10900k or even.

 

[arbiter]

A Con i think could added is outta box support. What i mean in that is if i buy a 5000 series and new board right now, there is a problem of will it work or am i boned if i don't have a pervious gen to drop in to the board for bios update. As much as i want a 5900x, i got a problem of if board was made recently enough to have new enough bios installed to support it when i don't have 1xxx/2xxx/3xxx to update bios if board doesn't have a way to do it without cpu as I would expect most cheaper boards don't.

 

[W1zzard]

I've said it before and I'll say it again, to all those wondering/arguing over TPU scores being lower vs. other reviewers: TPU tested with an older BIOS using AGESA 1.0.8.0, while others used the intended BIOS for Ryzen 5000, with AGESA 1.1.0.0 on their boards (AMD mentioned this in their videos).
Based on this, I tested it myself and it became clear there is a substantial difference between the two when running Cinebench R20 1T or games like SOTR.

Not true. I used the AMD recommended BIOS, which is AGESA 1.1.0.0. When I wrote 1.0.8.0 I didn't look up the actual value in the BIOS and just wrote what I thought it was.

More details here: https://www.techpowerup.com/forums/...0-series-tpu-review-discussion-thread.274274/

I'm making good progress with my investigation btw and will soon have surprising details for you.

 

[HTC]

I'm making good progress with my investigation btw and will soon have surprising details for you.

According to Gamers Nexus, 5600X get a boost in performance when using 4 sticks of RAM VS 2: the difference is between 2.5% to over 8% and is consistent.

Unsure if this also translates to other Zen3 CPUs.

 

[metalkhor]

Using RTX 2080 Ti that is out dated and lacks some key next gen features such as PCI-E Gen 4 makes this reviews as a whole irrelevant and misleading.
This is a big disappointment to see that TPU as a once leading publisher in tech world now become a weak and unreliable source of information.

 

[Deleted member 24505]

TPU as a once leading publisher in tech world now become a weak and unreliable source of information.

Says someone with 7 posts

 

[mainlate]

X570 needs more refreshed boards. 3900x soon below 300€, very good.

 

[avatar_raq]

I second what other people said, @W1zzard please reexamine the new CPUs with 4 sticks of RAM and see if you can reproduce the results of Steve from GamerNexus.

 

[Shou Miko]

I second what other people said, @W1zzard please reexamine the new CPUs with 4 sticks of RAM and see if you can reproduce the results of Steve from GamerNexus.

You mean the video Gamers Nexus released yesterday? It could be the reason why @W1zzard's Ryzen 5000 series scores ain't as high as others.

@avatar_raq but there are people who learned to use 2xsticks instead of 4xsticks because it's a dual-channel platform so it should be 2xsticks not 4xsticks since 4 sticks makes the mem controller run harder and can cause problems with stability even I haven't personally seen this for years but still can happen.

 

[bmacsys]

Didnt they say, that it will outperform 10900K in games too?

This seems like just another AMD launch, where the reality is completely different than the promises.

Oh my. Another weak attempt at trolling.

It's not an upgrade to Intel users lol.

You are too funny. I guess Covid-19 is fake too.

Intel used to release new substantially faster CPU architectures without doing this: Sandy Bridge, Haswell, Sky Lake were all a lot faster than previous generation CPUs without price hikes and in certain cases even cost substantially less than their predecessors, e.g. the Intel Core i5-2500K was released for $216 while the Intel Core i7-920 cost $305.

What's bad for Intel (charging top dollar) is absolutely OK for AMD because ... because AMD. We get it. Let's see if AMD keeps raising prices for the next Ryzen series (Zen 4/DDR5) and what new crappy excuses AMD fans will come up with.

Birdie/Yeshua the troll is everywhere. Copying and pasting the same bogus posts on dozens of different forums.

 

[Selaya]

No worries, often some nice improvements come out of feedback like this

Speaking of improvements, maybe mention that you're running custom instead of built-in game benchmarks? That seems to be the cause of some of the deviations between your and others' testing, and a custom benchmark would perform differently from a built-in one (and may be closer to the real world).

 

[okbuddy]

According to Gamers Nexus, 5600X get a boost in performance when using 4 sticks of RAM VS 2: the difference is between 2.5% to over 8% and is consistent.

Unsure if this also translates to other Zen3 CPUs.

hey, 4 sticks always better (zen 2 example)

source https://www.tomshardware.com/reviews/amd-ryzen-3000-best-memory-timings,6310-2.html

 

[mkppo]

This is exactly what I asked W1zzard, I knew there were performance gains to be had from 4 sticks (or dual rank 2 sticks). Zen 3 gains a lot more performance from quad ranks. Add in memory tuning and there's SO much more performance to be had from the configuration TPU used.

5600X looks faster than 10900K when 4 ranks are populated and 3600 mhz memory is used for both. 5800x is faster still. All reviews confirm this. 3600 sticks are close to 3200 prices, and 4 sticks of 3600 really allow Zen 3 to shine. You can still save a ton of cash compared to 10900K and simply get faster gaming performance.

 

[Calmmo]

Not true. I used the AMD recommended BIOS, which is AGESA 1.1.0.0. When I wrote 1.0.8.0 I didn't look up the actual value in the BIOS and just wrote what I thought it was.

More details here: https://www.techpowerup.com/forums/...0-series-tpu-review-discussion-thread.274274/

I'm making good progress with my investigation btw and will soon have surprising details for you.

But will you last enough to withstand the attack of the fanboys? is the real question The true phantom menace

 

[arbiter]

Using RTX 2080 Ti that is out dated and lacks some key next gen features such as PCI-E Gen 4 makes this reviews as a whole irrelevant and misleading.
This is a big disappointment to see that TPU as a once leading publisher in tech world now become a weak and unreliable source of information.

.... lacking pci4 doesn't make it irrelevant as they tested scaling and show that can use even pci2 16x and lose very little performance. Reason to use that card is cause they have a WHOLE HOST of results performance data to compare it to then just change it. They tested pci scaling using a 3080 and between 3.0 and 4.0 there was only around 1% loss.

NVIDIA GeForce RTX 3080 PCI-Express Scaling

NVIDIA Ampere finally brings PCI-Express 4.0 support to the high-end graphics market. The new interface promises twice the bandwidth of PCI-Express 3.0. We've setup an AMD Ryzen 3900XT system to test how various PCIe generations and lane widths affect gaming performance.

www.techpowerup.com

 

[Shatun_Bear]

It would be advisable if the whole set of reviews were re-done as this episode has been really damaging to TechPowerUp's reputation, and for my faith in their work unfortunately.

 

[arbiter]

It would be advisable if the whole set of reviews were re-done as this episode has been really damaging to TechPowerUp's reputation, and for my faith in their work unfortunately.

Testing is done in a way to make it apples to apples, when you use different memory which can be identical on bother sides now that makes question of is cpu faster or is memory unfairly bottlenecking 1 or the other. Both using as much of same hardware as possible removes that question of hardware like memory or gpu being the one doing it.

 

[Shatun_Bear]

Testing is done in a way to make it apples to apples, when you use different memory which can be identical on bother sides now that makes question of is cpu faster or is memory unfairly bottlenecking 1 or the other. Both using as much of same hardware as possible removes that question of hardware like memory or gpu being the one doing it.

I dont think it's just the memory that was the problem. The difference with other reviews is significant.

 

[Shou Miko]

What about 2x16GB kits?

In theory all 16GB sticks have memory chips on both sides of the PCB while almost if not all 8GB stick only have on one side of the PCB.

 

[okbuddy]

rumor: 6800xt samples already arrived, but the tech sites can't post anything