Best Eight-Core CPU Battle: AMD Ryzen 7 3800X vs Intel Core i7-9700K

Dan_D said:
You can compare these CPU's in a review as I have done so in the past. I never compared them directly, but I did include numbers for the 9700K in the lineup.
Click to expand...

I'm not defending the article from OP, or attacking your reasoning here Dan_D but as I lurk around the internet, I don't see the same logic applied when comparing AMD & nVidia GPUs and I find that both curious and disappointing.
This was to be a short two or three sentence post, but the more I thought about it characters just kept popping in. :ROFLMAO:

Do we take the same approach when comparing similar priced AMD & RTX nVidia cards?
How about we forget about RTX for a moment, how about the hardware differences between the two architectures from even a high level, CUDA vs stream processors?
Completely different designs that both produce graphics for a consumer.

Based on reasoning in this thread, we should be defending AMD against RTX because their hardware isn't built the same and lacks certain features, so obviously their performance without these features is going to be less than nVidia RTX products?

Where is the line drawn, exactly?
Is it AMD's fault they are able to sell a product with twice as many threads at a similar price point as an Intel product?
If Intel quadruples their core/thread count tomorrow and sells at the same price as AMD, does AMD receive this much defense because obviously, comparisons are not going to be fair?

That's literally good business. Being able to offer a 'better' performing product at similar or lower price to consumers.

Price is the factor to a lot of people, it's the most valid and relevant comparison one can make.
However, it's up to the consumer to make that comparison and those reviewing the products shouldn't do it for them.
But, how then does one compare CPUs when use case scenarios are so varied and there's no one good metric to use?

It's newsworthy, and yeah the Tom's article kinda reads like a reach around for AMD, but after all the years of the roles being reversed, it's only fair. (up to a point!)
 
Meh, its a moot point. The only way the 9700k is a good deal is if you get in on the microcenter deals at $275-299 tops and you have to go intel for single core performance in the software/games you use. Its been this way since the 9700k's release. I don't even need to bring AMD into the comparison, just look at the history of i5's vs i7's with hyperthreading. Long term viability of a 8c8t chip is questionable. Those 4c4t 7600 and even the 6c/6t 8600 chips are struggling in new games for the past year. It will only get worse. Hence why Intel is bringing back hyperthreading to most of its lineup this year. As good as my sandy bridge 2500k was, it was obvious by haswell and certainly skylake that the i5 was a bad buy for long term 3+ years viability across all usage types. With the explosion in cores and threads, its only going to get worse. Bring back in AMD and look at how well the r5 1600 and 2600's hold up over the last 2-3 years vs i5's and its no contest.
 
Inacurate said:
I'm not defending the article from OP, or attacking your reasoning here Dan_D but as I lurk around the internet, I don't see the same logic applied when comparing AMD & nVidia GPUs and I find that both curious and disappointing.
This was to be a short two or three sentence post, but the more I thought about it characters just kept popping in. :ROFLMAO:

Do we take the same approach when comparing similar priced AMD & RTX nVidia cards?
How about we forget about RTX for a moment, how about the hardware differences between the two architectures from even a high level, CUDA vs stream processors?
Completely different designs that both produce graphics for a consumer.

Based on reasoning in this thread, we should be defending AMD against RTX because their hardware isn't built the same and lacks certain features, so obviously their performance without these features is going to be less than nVidia RTX products?

Where is the line drawn, exactly?
Is it AMD's fault they are able to sell a product with twice as many threads at a similar price point as an Intel product?
If Intel quadruples their core/thread count tomorrow and sells at the same price as AMD, does AMD receive this much defense because obviously, comparisons are not going to be fair?

That's literally good business. Being able to offer a 'better' performing product at similar or lower price to consumers.

Price is the factor to a lot of people, it's the most valid and relevant comparison one can make.
However, it's up to the consumer to make that comparison and those reviewing the products shouldn't do it for them.
But, how then does one compare CPUs when use case scenarios are so varied and there's no one good metric to use?

It's newsworthy, and yeah the Tom's article kinda reads like a reach around for AMD, but after all the years of the roles being reversed, it's only fair. (up to a point!)
Click to expand...

I think the issue here, is that I wasn't terribly clear about the price comparison. Off hand, I hadn't really looked at their pricing. I just knew they were kind of close. Upon checking them, the 3800X and 9700K are anywhere between $30 and $50 apart. I think this pretty clearly puts them in the same price point. Having said that, there are numerous ways to compare a CPU. There are many reasons to compare CPU's of varying price points, performance levels, and even compare hardware from different segments. However, this is where we get into the nitty gritty of the review's actual purpose.

Largely, these days I have quite a bit of decision making power over what content I produce vs. what's assigned to me. This hasn't always been the case. When I worked for HardOCP, I made absolutely no decisions about the content I produced. I could write an editorial, but that never guaranteed a release. I was provided hardware and while I could ask for things I felt I needed, and give input on the article in terms of what I thought should or shouldn't be in it, I still didn't have much of a choice. I bring this up for reasons I'll get into shortly.

The purpose of a review is to generate page hits which ultimately, mean revenue. I think we can all agree that this is the purpose of any hardware review. There are many reasons to choose a particular subject matter, comparison or to review a specific piece of hardware. Reviews are often about timing. For example, I reviewed the ASUS Prime X299 Edition 30. Now, ordinarily, I wouldn't have reviewed this motherboard from a business perspective, because it's late in the X299 life cycle. However, as luck would have it, Intel released it's 109xx series Cascade Lake-X family of CPU's and thus, X299 became a little more relevant again. ASUS had this motherboard on hand and sent it to me to support the 10980XE article and to get a review out of it.

I reviewed this motherboard largely because I had gathered most of the testing data from doing the Core i9 10980XE CPU review anyway. So, it wasn't much of a stretch to go ahead and finish a full review of the motherboard itself. Again, X299 has slightly more relevance now, but I wouldn't review any more X299 motherboards as the window for it has pretty much already passed. I have at least one more X299 motherboard on hand, but there just isn't much point to it. Most of the people that will buy into X299 already have. It turned out to give us a bit more exposure than I'd have imagined, so reviewing this motherboard was good for us. Primarily because, no one else really reviewed it.

Getting back to the Tom's Hardware article. From a review perspective, it seems weird to make this comparison now. To be clear, I have no problem with the fact that the article pits a 9700K against a 3800X based on their respective price point. But, the article's text focuses only on those two despite providing data samples for a wider range of applications and games. They show data for a fairly large segment of processors from both sides in 9 different games and some GeoMean test suite, (whatever that is). That all actually looks good.

Where I have the real problem with the article, is the bias seen at the end which reads like the author is looking to bend over for a pegging session with Lisa Su. By that I mean, I disagree with conclusion when the author states: "We don't expect anyone to pick up a Core i7-9700K just to play games, so gaming performance ultimately takes the backseat." This is precisely the opposite of why someone would pick up a 9700K. The only reason to get one is because you want it for gaming. Thus, the author downplays the gaming performance of the processor, in spite of the fact that the article's gaming benchmarks show the 9700K at 5.1GHz dominating every single CPU in the lineup most of the time. The article acts as though the gaming performance difference didn't exist and made it sound like there was no reason to buy a 9700K.

Even the price point comparisons between the two platforms are myopic. The article presumes that you are buying everything for a new build. It doesn't account for people who already have a Z390 or other compatible motherboard and just need a CPU. In that case, if you are just a gamer or have a more limited budget, the 9700K could be a good fit for you. Anything that could put the Intel CPU in a positive light was purposefully ignored or at best downplayed.

There are lots of reasons not to buy a 9700K, but the article doesn't really talk about anything but price and multi-threaded workloads. A comparison that is only fair by price point alone. I spoke off the cuff when I said the 3800X vs. 9700K comparison was ridiculous. It isn't in the price point perspective or even as pure entertainment. However, I do stand by the statement simply because the article doesn't really serve a purpose given that the outcome of such a test was already well known. It doesn't make much sense to do such an article, and when you look at what little is in it, you can kind of see it was just put together because it was easy.

It reads like something from the National Enquirer rather than a reputable and unbiased review site. Now, the author may not be to blame for this. I hate to defend Tom's Hardware, but it may not have been the author's choice. Similarly, Tom's did a lame VRM test on MSI's low end boards that MSI admitted weren't that great, and Tom's concluded the boards were fine and everyone else was wrong. The author of that article only used a 3700X to test those boards rather than a 3950X. The fact of the matter is, that may be all the author was given to test with. These companies tend to only send out one or two examples of a CPU, and not all the reviewers will have full access to what they should have.

Of course, the way that article was written, it doesn't tackle that issue and instead draws a false conclusion based on a limited testing setup and knows that it goes against the grain of what other reputable sites have shown, but instead of acknowledging that, they toss whatever integrity they had out the window by calling other sites "liars." So again, when we see a AMD fanboy circle-jerk from a review site, people call them out for it. It's really not that the comparison is unfair in a technical sense, but rather that data was blatantly ignored and the full picture of the comparison was distorted to fit a specific narrative.

It is hard to be upset about Intel being "mistreated" by an articel given some of what Intel has done in the past and given that AMD's been under it's heel for so long. However, bias is bias, even when that bias aligns with the truth. The 3800X is a better buy, but it completely ignores the potential reasons why a 9700K might be a good fit for someone or what it's strengths were.
 
Dan_D said:
I'll tell you the same thing I tell everyone else. If your happy with your current gaming experience, then it doesn't make much sense to upgrade. If you aren't, then an upgrade is perhaps warranted. I know people ask questions like this because they want new hardware and aren't sure what to expect from teh change. The Ryzen 7 1700 isn't a great gaming CPU. It just isn't. It was well behind Intel's CPU's from several years ago on the gaming front. You would see quite an improvement upgrading to a 3600 but almost none going to a Ryzen 7 2600. There is a big change going from a 1st or 2nd generation Ryzen to the 3000 series. Especially where gaming is concerned. A Ryzen 5 3600 or 3600X would be a big upgrade on that front over a 1700.

As for a Vega 64, I know they are faster than the Vega56, but I am not really sure by how much. To be perfectly honest, if it isn't one of the top two graphics cards on the market, I don't pay too much attention to them. That said, a quick check on the subject seems to show that the Vega 64 isn't that much faster than the Vega 56 most of the time. However, the Radeon 5700 looks to be an alright upgrade over the Vega 56. However, I think that one depends on what resolution and quality settings your at. Sometimes it only seems to put out about 5-10FPS more, but sometimes that's in a game where the Vega 56 is running sub-60FPS. Those numbers don't sound like much but that's a pretty dramatic impact on your experience. Being at 49FPS and cracking 60FPS is a pretty big jump in performance. In other cases where you might be seeing 89FPS and jumping to 101FPS, it isn't as big a deal. 1%'s don't look much different between the two, but your averages do.

But again, that's a personal choice. Performance wise I say yes, but price is up to you. I'm the guy who buys $1,000+ CPU's and $1,200+ video cards. Naturally, I'm going to say yes as I can afford it. Where I tend to not take the upgrade is where the difference is well below 10% and the price is 40-50% more to get there.



It sure seems that way. As I said, it reads like someone is trying to fluff AMD for a porno shoot.




It's certainly spiraled downhill over the years. That's what happens when a media conglomerate buys your site and all it gives a shit about is ad revenue.
Click to expand...

Oops, sorry, I meant that I have those three setups actively running at this time. :) ;) For running at 1080p 144hz with Freesync, I will probably stick with the 1700 just a little while longer, or until I receive my tax refund. :) You are correct though, I always seem to get the upgrade bug every now and then but, I do not need it and can wait a little while longer. I just love windows shopping but, I do not need to spend the money.
 
Master_shake_ said:
i think the 9700k being more expensive is the real thing wrong here.
Click to expand...
If Intel doesn't lower prices, then people will simply not buy them except when they're looking for the specific advantages they bring...

It's not like Intel doesn't know that they could lower prices to increase sales, after all ;).
 
Master_shake_ said:
well the 3700x is an 8 core cpu.

adding the suffix "thread" is a new thing.

we didn't call a q6600 4 core 4 threaded cpu now did we?
Click to expand...
We didn't use the 'thread' suffix because before Zen, AMD didn't have SMT. It was simply 'HT' added to the CPU name, at first, and for a while Intel didn't use HT at all. For AMD, Bulldozer was an ill-fated attempt at 'hardware SMT', but that can only be argued to be a form of SMT because AMD had to pay out for those extra execution units not being equal to more cores ;).

So yeah, with Zen, we specify number of cores and threads together, whether we're talking about x86 CPUs or ARM or IBM Power CPUs with four threads per core.
 
Dan_D said:
fluff amd for a porno shoot
Click to expand...

0CDF6D45-D4AC-421E-89FC-6C0E3E081C84.jpeg

Sorry Dan, your comment made me remember this 🍺
 
Last edited:
Krazy925 said:
I can’t believe toms would compare 2 CPUs in the same price range. Lulz.
Click to expand...
To Dan's point though you wouldn't compare a $30,000 Camry and a $30,000 Subaru BRZ and claim the Camry is a better car because it seats more people. While that is true, it ignores the reason people buy a BRZ.

If you review both CPUs and ignore the gaming aspect of the 9700K - which is it's only actual strength at this price - you're abitrarily setting up the CPU to fail.

And I own two Ryzen systems so I am not biased at all in this respect.
 
Rizen said:
To Dan's point though you wouldn't compare a $30,000 Camry and a $30,000 Subaru BRZ and claim the Camry is a better car because it seats more people. While that is true, it ignores the reason people buy a BRZ.

If you review both CPUs and ignore the gaming aspect of the 9700K - which is it's only actual strength at this price - you're abitrarily setting up the CPU to fail.

And I own two Ryzen systems so I am not biased at all in this respect.
Click to expand...
Eh I mean yes and no right. You would compare the utility of the money you are spending. I agree with your point about seats of course but at the end of the day we have a finite amount of resources. The BRZ definitely doesn't fit as many use cases as Camry, which is why it sells many less.

Most consumers in both these spaces make an either or decision. AMD or Intel (and $300 seems to be close to the sweet spot (maybe a little high even)) for processors in your average "gaming rig". Similarly cars, most people don't own multiple cars. The need is satiated by the purchase of the other car.

I could babble on about the utility of money for hours, but ain't nobody around here got time for that.

Toms sucks and the article isn’t good but people know what they want from their processors. Let’s toms do the stupid reviews. The 9700k is good at what it’s good at and Ryzen is a great challenger. Be fair though Ryzen does excel in many areas and that seems to bring some rage for some reason.

PS I flip back and forth between AMD and Nvidia and Intel depending on my needs, I'm not gonna accuse anyone of being a fanboy, because frankly I'm not super invested in this whole debate or argument anyways. At the end of the day I buy whatever provides the most value or use for the money I want to spend.
 
It's okay most people here don't understand why cpu gaming tests are ran at low resolutions.
 
Jim Kim said:
The 3800x is a joke unless it is priced beneath a 3700x.
don’t believe me, just ask digital jesus
Click to expand...

The 3800X is a better binned part plain and simple. If you plan to actually overclock to reach maximum performance potential it will reach higher clocks on average than a 3700X.
 
FlawleZ said:
The 3800X is a better binned part plain and simple. If you plan to actually overclock to reach maximum performance potential it will reach higher clocks on average than a 3700X.
Click to expand...
The average Joe won't see a difference. The chip has already seen deep sales because it ain't selling. Watch the video, read the reviews. It's a dud.
 
  • Like
Reactions: Auer
like this
Given more people place more weight on price in the factor of buying a CPU than other things, then I feel that would be the most valid comparison.
For instance my buying preferences I might consider: 51% price, 45% performance, and the rest being other things like power usage or efficiency. So a 3700x/3800x might score higher in that criteria.
I agree that Tom's article sucks balls, but as a reviewing strategy (directly competing based on price) it makes the most sense.
I feel like that's also what HardOCP has done throughout its history, a CPU/GPU would have to be a certain amount faster for it to be "worth it" over a slower and less expensive one. I mean isn't that why the 4870 was so popular back in the day?
 
Jim Kim said:
The average Joe won't see a difference. The chip has already seen deep sales because it ain't selling. Watch the video, read the reviews. It's a dud.
Click to expand...

Oh, so the 1600, 2600, 2600X, 2700, 2700X, 3600 and 3600X ain't selling either then, eh? No, it is not a dud, specially at the sale prices but, at MSRP, it was not worth it. However, the 3800X is a better binned part and the 1800X was as well. No, the 1700 and 1700X did not equally overclock as well as the 1800X but, that was fine, it was still worth it, at the time. (I did not buy one but wish I had, anyways.)
 
The review is still relevant. I just built a budget machine for a client. They had 300ish to spend on a cpu, do 0 gaming, do a mix of single and multi threaded workloads with heavy multitasking but no rendering. I set them up with a 3800x (which does boost a bit more than my 3700x did regularly), a b450 motherboard (which has regular bios updates as recent as a month ago), and filled 2 of the dimm slots with 2x8gb 3200mhz. Total build cost with case, nvme, gtx 950, came out to about 550 I think. The motherboard is basic, but it supports upgrading to the 3950x in the future, and hopefully next gen ryzen.

My point is what has been said in this thread already. For a given price point, unless you have a specific need (gaming), the amd system is mostly equal or better. No fanboyism here, I just like cheap performance and that's what amd is serving right now.
 
ManofGod said:
Oh, so the ... snip
Click to expand...
If the only place anyone can °feel° a difference is with benchmarks than it's a dud in my book.
No one in this thread could feel, see or smell the difference between an over locked 3800x and a normally aspirated 3700x.
 
Yes, the 9700K is 5-10% faster in games. That is really the only thing it is better in. The 3800x is cheaper and it comes with a cooler. I agree you shouldn’t discount games, but overall, the 3800x seems to be the better processor. Besides, if l was buying a processor strictly for gaming right now, I’d probably get the 3600 as it has similar performance for less money. And it still has more threads than the 9700K.
 
tdkuhl said:
Yes, the 9700K is 5-10% faster in games. That is really the only thing it is better in. The 3800x is cheaper and it comes with a cooler. I agree you shouldn’t discount games, but overall, the 3800x seems to be the better processor. Besides, if l was buying a processor strictly for gaming right now, I’d probably get the 3600 as it has similar performance for less money. And it still has more threads than the 9700K.
Click to expand...

Yeah, from what I am seeing, unless you are running a 2080Ti, you will not benefit much at all from a 9700K. Of course, if you have a 2080ti, the 3900X or 9900K is probably the go to processor, at that point.
 
ManofGod said:
Yeah, from what I am seeing, unless you are running a 2080Ti, you will not benefit much at all from a 9700K. Of course, if you have a 2080ti, the 3900X or 9900K is probably the go to processor, at that point.
Click to expand...
Totally agree there. If you have to have absolute best frame rates, you’re shelling out the money for best performance bar none.
 
tdkuhl said:
Yes, the 9700K is 5-10% faster in games. That is really the only thing it is better in. The 3800x is cheaper and it comes with a cooler. I agree you shouldn’t discount games, but overall, the 3800x seems to be the better processor. Besides, if l was buying a processor strictly for gaming right now, I’d probably get the 3600 as it has similar performance for less money. And it still has more threads than the 9700K.
Click to expand...

The difference in gaming shouldn't be undersold. The Intel CPU's are roughly 6% faster in aggregate but can be as much as 15% faster or more in some isolated cases. You really need to look at the games you play. Initially, when AMD first got the 3000 series working on Destiny 2, they lagged their Intel counterparts by a much wider margin than that. This has since been corrected, but for a long time the Intel CPU's provided a substantially better gaming experience.

This is still the case in both Hitman 2 and Shadow of the Tomb Raider:

1582905042208.png


1582905053892.png


These are isolated examples where the frame rate improvement can be substantially higher on Intel CPU's going well beyond the usual 5-10% increase. That gap widens a bit if your lucky enough to get an Intel CPU to 5.1GHz.

Having said all of that, we are still splitting hairs at 1920x1080. At higher resolutions the data is slightly different. The gap is still there, but its somewhat less impactful unless you drop into the lower price points for CPU's.

1582905289122.png


Here you can see the Ryzen 2000 series taking a massive hit compared to their 3000 series counterparts. Also, the Ryzen 5 3400G isn't all that high end and gets trounced by the Ryzen 7 2700X. The 9900K has a distinct advantage here.

Here we can see the Ryzen 9 3900X struggle to maintain a 60FPS minimum that an Intel HEDT CPU can actually maintain consistently. What I've seen in some games on the AMD side was true here as well. The maximum FPS can often be higher than it is on Intel machines, but for some reason the lows are much worse. Usually the averages are the same or better than Intel's. Part of the problem, and the reason I've hated testing with the Ryzen 3000 series is that the AGESA updates and inconsistent boost clocking put their results all over the map.

1582905475258.png
 
Last edited:
Dan_D said:
The difference in gaming shouldn't be undersold. The Intel CPU's are roughly 6% faster in aggregate but can be as much as 15% faster or more in some isolated cases. You really need to look at the games you play. Initially, when AMD first got the 3000 series working on Destiny 2, they lagged their Intel counterparts by a much wider margin than that. This has since been corrected, but for a long time the Intel CPU's provided a substantially better gaming experience.

This is still the case in both Hitman 2 and Shadow of the Tomb Raider:

View attachment 226610

View attachment 226611

These are isolated examples where the frame rate improvement can be substantially higher on Intel CPU's going well beyond the usual 5-10% increase. That gap widens a bit if your lucky enough to get an Intel CPU to 5.1GHz.

Having said all of that, we are still splitting hairs at 1920x1080. At higher resolutions the data is slightly different. The gap is still there, but its somewhat less impactful unless you drop into the lower price points for CPU's.

View attachment 226612

Here you can see the Ryzen 2000 series taking a massive hit compared to their 3000 series counterparts. Also, the Ryzen 5 3400G isn't all that high end and gets trounced by the Ryzen 7 2700X. The 9900K has a distinct advantage here.

Here we can see the Ryzen 9 3900X struggle to maintain a 60FPS minimum that an Intel HEDT CPU can actually maintain consistently. What I've seen in some games on the AMD side was true here as well. The maximum FPS can often be higher than it is on Intel machines, but for some reason the lows are much worse. Usually the averages are the same or better than Intel's. Part of the problem, and the reason I've hated testing with the Ryzen 3000 series is that the AGESA updates and inconsistent boost clocking put their results all over the map.

View attachment 226613
Click to expand...
All of those are 1080p Low settings. What were the differences on medium and high? What GPU was used?
 
1582907105947.png


1582907207337.png

Here are some comparisons with different video cards as the settings quality goes up and the graphics card goes down, the numbers even out. This is also where I think spending less on a processor and more on a video card would be beneficial.
 
tdkuhl said:
All of those are 1080p Low settings. What were the differences on medium and high? What GPU was used?
Click to expand...

I never tested these at medium settings and I only have data at high settings for a couple of games. Unfortunately, being a new site I didn't have a display that could do 4K on the test bench initially. So, earlier CPU's didn't include those settings. Not only that, we got review samples incredibly close to the launch date and were crunched for time. Unfortunately, some planned testing didn't make it into the final articles.

However, I do have a fair amount of data for Destiny 2. I started pulling data for Ghost Recon Breakpoint, but most of it hasn't been included in any reviews yet. I'm still tweaking the formats of my articles and deciding what to do, where to do it, etc.

For example:
1582907851433.png

This is between a 3900X and a 9900K. The GPU used in most of the reviews is a GIGABYTE RTX 2080 Ti Aorus Xtreme 11G. That is my personal video card and after we got enough review samples in it was replaced with a MSI RTX 2080 Super X-Trio Gaming.
 
People seem to be bothered that the 9700K is currently an excellent option for gamers with a healthy budget. Games ported from PS5/XBSX will likely change that, so I bought a 3900X to prepare for that era. But it currently doesn't do anything extra for me. I'm not creative and won't ever stream anything, so I would have been very happy with a 9700K.
 
Jim Kim said:
The average Joe won't see a difference. The chip has already seen deep sales because it ain't selling. Watch the video, read the reviews. It's a dud.
Click to expand...
I've seen the reviews and videos. I own one. I didn't buy the 3800X for it's out of the box experience. I bought the 3800X to have the best 8 core Ryzen 3000 for pushing maximum performance. And I did this with a mere $30 price delta over the 3700X. My 3800X can do 4.5Ghz all core in games and benches all day long. One core will fail after running prime for a while but I don't play prime95. I don't see any 3700X doing that.
 
tdkuhl said:
Here are some comparisons with different video cards as the settings quality goes up and the graphics card goes down, the numbers even out. This is also where I think spending less on a processor and more on a video card would be beneficial.
Click to expand...
Really depends on when you plan on upgrading either part. That's the reason that '1080p Low' results are important for CPU reviews and why reviewers don't just test 4k60 for CPU reviews and be done with it.

You want to see how CPUs scale. The IPC and clockspeed ceiling mean that Ryzen CPUs scale a bit less.
 
FlawleZ said:
I've seen the reviews and videos. I own one. I didn't buy the 3800X for it's out of the box experience. I bought the 3800X to have the best 8 core Ryzen 3000 for pushing maximum performance. And I did this with a mere $30 price delta over the 3700X. My 3800X can do 4.5Ghz all core in games and benches all day long. One core will fail after running prime for a while but I don't play prime95. I don't see any 3700X doing that.
Click to expand...
So you don’t want a stable computer. O.K. https://xkcd.com/2250/
I’m the total opposite.
 
I think the biggest mistake Intel has done is to stubbornly keep the prices of their CPUs so high, the 9700K should have been $339 MSRP by now. I don't even believe it's supply & demand or shipping & manufacturing issues in this case. It's just stubborness in admitting "defeat" from the administration.

I had thought about swapping out the 8600K to 9700K when it gets cheap, guess the "cheap" never came and probably never will, they rather just see it quickly EOL disappear at this point when the new series have been out for a while.
 
Last edited:
I agree it is a good gaming CPU. I just don’t think it is good value. $339 for the 3800X vs $396 for the 9700K.
 
FlawleZ said:
Stability is subjective. I want my desktop computer fo turn on everytime and play or do whatever I want to throw at it without a hiccup. And it does just that.
Click to expand...

No.
There is Stablility, when all tested parts of the CPU perform without fault under a true 100% load (synthetic usually) indefinitely , and "Stable enough" which is what you described.
Do not confuse the two.

That said.... My 1700 is "Stable enough" at 4GHz, tested with quick burn tests with IBT, Prime95, Linpack, realbench. But I am not going to run my 24+ hour x265 workloads at that speed. (I run them at stock speeds)
I don't feel that running the encodes @ 4GHz is worth the possible: BSOD, corrupt output file, strained Mobo, shortened life, etc, for a slightly faster encode time. It'll probably finish while I'm not at the computer anyways...
But...... I like running my games at 4GHz
 
Last edited:
Rockenrooster said:
No.
There is Stablility, when all tested parts of the CPU perform without fault under a true 100% load (synthetic usually) indefinitely , and "Stable enough" which is what you described.
Do not confuse the two.
Click to expand...
I get pushing hardware, and the joy out of getting more out of it -- but I also don't enjoy wondering if or when a system I'm using seriously (work or gaming or whatever) will crash. I agree with the difference that you've noted here. There's 'it works for now' and then there's actually stable.
 
tdkuhl said:
I agree it is a good gaming CPU. I just don’t think it is good value. $339 for the 3800X vs $396 for the 9700K.
Click to expand...

I hear you. $90 difference in Canada at moment, PLUS a half decent air cooler for the 9700K which is about $30 CAD.

But there are always people willing to pay dearly for the increase of the tiniest performance metrics.

Also, "priced higher = better" is a fallacy that marketing know people still fall for, despite the fact conflicting evidence is just a google search away.
 
VIC-20 said:
I hear you. $90 difference in Canada at moment, PLUS a half decent air cooler for the 9700K which is about $30 CAD.
Click to expand...
Given that the AMD parts don't really overclock and the Intel parts do -- a better cooler for AMD is likely to only get you a little less noise under load, while it'll actually get you better performance with Intel.

Or at least that's an angle that makes a bit of sense. AMDs coolers have been pretty impressive and spot on for the required performance; Intel could offer the same, certainly, but never has for K-series CPUs. Of course, that also makes a bit of sense too, as you're paying up a bit for the performance (when it matters), and likely overclocking too, so purchasers are likely to want something better and be willing to pay for it.
 
You must log in or register to reply here.
Back
Top