Ryzen 5 3500 6c/6t possibly coming

[P4-630]

https://twitter.com/i/web/status/1164711479662141441

Ryzen 5 3500 Specs Leak: Could Be New Budget CPU King

The Ryzen 5 3500 is a hexa-core Zen 2 processor that's rumored to come without SMT.

www.tomshardware.com

 

[FinneousPJ]

Interesting. I wonder how it will compare to the 3600 non X in price to perf.

 

[EarthDog]

Those clocks for a 6c/6t part though...

 

[Bill_Bright]

Those clocks for a 6c/6t part though...

IDK. Compared to its [supposedly] direct competition, the i5-9400, the Ryzen 5 3500 has:

• An anticipated competitive price somewhere between $150 - $185,
• Faster DDR4 support,
• PCIe 4.0 support vs PCIe 3.0 for the i5,
• 3.6 times larger cache,
• And a faster base of 3.6GHz vs only 2.9GHz for the i5.

So I don't have a problem with the Ryzen 5 3500's clocks, even if the boost is the same - assuming those "unconfirmed" specs hold.

 

[MrGRiMv25]

Interesting move from AMD considering pretty much all their parts have SMT enabled. Should give the 9400 a good fight if price is decent, if I were to get a Ryzen I'd still get the SMT enabled version but still seems like a decent product nonetheless.

 

[Deleted member 67555]

The 3500 would have to be $130 or less to be competitive with the 9400f.
They might do a 3500x but I bet everything lower will be for HP and Dell.

 

[Bill_Bright]

The 3500 would have to be $130 or less to be competitive with the 9400f.

No one is comparing the 3500 to a 9400F. It was comparing it to the 9400. They are not the same processor.

 

[XL-R8R]

No one is comparing the 3500 to a 9400F. It was comparing it to the 9400. They are not the same processor.

The 9400F is just lacking iGPU as you probably already know; however the 9400F - at least in my area of the world - is being sold for significantly less than the 9400.


But, I still would take the 3500. I also agree with the points made in an earlier @Bill_Bright post - the AM4 platform offers more value for money in this segment either which way you look at things.

 

[Deleted member 67555]

No one is comparing the 3500 to a 9400F. It was comparing it to the 9400. They are not the same processor.

I know but since the -f is without an Igpu it compares better than the regular 9400 against the 3500...
I would probably take the 3500 as well simply for the features you pointed out in your first post... And because it's on a live platform

I could see these going for $130-$150.

 

[EarthDog]

IDK. Compared to its [supposedly] direct competition, the i5-9400, the Ryzen 5 3500 has:

• An anticipated competitive price somewhere between $150 - $185,
• Faster DDR4 support,
• PCIe 4.0 support vs PCIe 3.0 for the i5,
• 3.6 times larger cache,
• And a faster base of 3.6GHz vs only 2.9GHz for the i5.

So I don't have a problem with the Ryzen 5 3500's clocks, even if the boost is the same - assuming those "unconfirmed" specs hold.

pcie 4.0... useless.
Faster ddr4 base spec...useless...all part of its end performance value.
Cache...yippee...already included with performance...useless metric here.
Clocks... the Intel chips will boost similarly on all c/t and have dual cores and single faster.

Regarding clocks, the 9400/9400f has an all core boost to 3.9ghz. 2/4= 4.0 ghz, and one core is 4.1 ghz. AMD doesnt even hit its listed boost clocks for many people...Intel always will.

Price is about all it may have... if you're buying budget parts like this pcie bandwidth wont be an issue...you can still run 2 m.2 drives and have plenty of sata left over.

9400/9400f... same thing, one doesnt have an igpu... like the ryzen. So you could say that if one doesnt need a discrete gpu, the Intel is indeed a better buy.

 

[notb]

No one is comparing the 3500 to a 9400F. It was comparing it to the 9400. They are not the same processor.

But you should. That's the whole point of F processors.
9400F will likely be cheaper, easier to get and already an established product.
Performance should be more or less identical.

All the technical details you've mentioned are irrelevant. This is a mainstream gaming / office CPU. No one will care about PCIe 4.0 and cache size. Target group looks at performance, not shiny specs.

 

[Vayra86]

This is probably a very poor bin, they need the voltage for 6 cores and can't reliably enable SMT, and the clocks are low.

For gaming I would avoid this one (just as I would the 9400(F)...). It will feel old fast. Better off upping the game just a little bit to get SMT and a few hundred more mhz.

The 'cheap i5' route doesn't work like it used to back in Sandy/Ivy/Haswell days. Progress is on now, and core counts increase faster than they ever did. Thread count headroom is something you'll want.

 

[Lionheart]

pcie 4.0... useless.
Faster ddr4 base spec...useless...all part of its end performance value.
Cache...yippee...already included with performance...useless metric here.
Clocks... the Intel chips will boost similarly on all c/t and have dual cores and single faster.

Regarding clocks, the 9400/9400f has an all core boost to 3.9ghz. 2/4= 4.0 ghz, and one core is 4.1 ghz. AMD doesnt even hit its listed boost clocks for many people...Intel always will.

Price is about all it may have... if you're buying budget parts like this pcie bandwidth wont be an issue...you can still run 2 m.2 drives and have plenty of sata left over.

9400/9400f... same thing, one doesnt have an igpu... like the ryzen. So you could say that if one doesnt need a discrete gpu, the Intel is indeed a better buy.

Umm no you don't know the pricing or the performance of the 3500 so how is Intel's 9400/9400F a better buy?

 

[notb]

Umm no you don't know the pricing or the performance of the 3500 so how is Intel's 9400/9400F a better buy?

Performance of Zen is very easy to estimate. It's a unified architecture.

Price... well, surely cheaper than 3600 ($200), but by how much?
Zen2 is likely expensive to manufacture. They can't go too low.

It's almost certain they won't go below $150 - the current price of 9400F.
So 2 CPUs with same performance and price => no reason to buy AMD.

It's easy to notice that AMD asks a bit too much for these CPUs and they're trying to widen the target group.
Not so long ago AMD's fans praised them for not holding tech back - often specifically for giving all Ryzens SMT. No locking of features like the bad Intel does. ;-)

 

[FinneousPJ]

So 2 CPUs with same performance and price => no reason to buy AMD.

What silliness. You don't know their performance or price. And furthermore, you could as well say no reason to buy Intel.

 

[EarthDog]

Umm no you don't know the pricing or the performance of the 3500 so how is Intel's 9400/9400F a better buy?

as was said, its performance is going to be just as every other ryzen cpu is comparatively to the intel counterpart. So, slightly behind in gaming and similar in all c/t behavior. This time though, amd's smt efficiency advantage is gone. If it's priced $30-50 less and dont need a discrete GPU, that cost difference goes to getting a discrete GPU to throw an image on the screen.

That said if they are priced similarly at 150 and it's a gaming rig, I'd go intel. You can also lock all cores to the boost value, and perhaps get another 100mhz with bclk. I'm guessing these are potato overclockers like the rest of ryzen 3 and wont reach its boost clocks with all cores.

Either system will perform well, but if it's a gamer, intel will have the advantage at 1080p however slight.

 

[Vario]

Thats just not all that interesting a CPU unless they sell it for cheap, around $130 would suffice.

I've actually seen an i5 8400 on local classifieds for $100 and they routinely sell for around $120 used or open box. I wouldn't hesitate to buy a used i5 8400 because theres very little that can go wrong with a locked Intel processor.

 

[lexluthermiester]

This CPU sku was not a far stretch to predict. However upon release it would indeed become the new budget build sweet-spot.

 

[notb]

What silliness. You don't know their performance or price.

I don't know the price, true. But we know that the CPU above this one costs $199 and that AMD will be unwilling to go very low. 7nm is expensive and shareholders expect AMD to finally make money.
Some people here hope for ~$130. IMO that's unlikely. I expect $149 or $159 - to compete with 9400F ($150), which - obviously - will be replaced fairly soon.

However, performance is easy to estimate. We know how other Zen2 CPUs perform - also with SMT turned off. It should match 9400F very well.

And furthermore, you could as well say no reason to buy Intel.

Well, not exactly. Intel is the default choice. It has better brand, better availability, better OEM support and better marketing.
AMD has to offer something extra to gain traction: price, performance, better stock cooler, bundled games etc.

That said, 9400(F) is a CPU made for home desktops, which makes it a very niche product - not that important for Intel.
Much like with earlier generations, the _500 (9500, 9500T) is the i5 that Intel wants to sell in large quantities.
So if AMD decides to go for very aggressive pricing in this segment, Intel might let them have it.

 

[Solid State Soul ( SSS )]

The 3500 would have to be $130 or less to be competitive with the 9400f.

Totally

 

[Bill_Bright]

But you should. That's the whole point of F processors.

What I should or should not do is immaterial and not the point. The point is about the article/comparison the OP linked to and it used the 9400, not the 9400F. If you feel Tom's Hardware should have used the F processors instead, your argument should be with them, not me.

As far as the other specs being "useless" well all I can say there is "yeah right".

When comparing any two similar, nearly identically matched products, it is about all of the minute specification details that differentiate (or compare similarly) the two. Not just one.

 

[FinneousPJ]

I don't know the price, true. But we know that the CPU above this one costs $199 and that AMD will be unwilling to go very low. 7nm is expensive and shareholders expect AMD to finally make money.
Some people here hope for ~$130. IMO that's unlikely. I expect $149 or $159 - to compete with 9400F ($150), which - obviously - will be replaced fairly soon.

However, performance is easy to estimate. We know how other Zen2 CPUs perform - also with SMT turned off. It should match 9400F very well.

Well, not exactly. Intel is the default choice. It has better brand, better availability, better OEM support and better marketing.
AMD has to offer something extra to gain traction: price, performance, better stock cooler, bundled games etc.

That said, 9400(F) is a CPU made for home desktops, which makes it a very niche product - not that important for Intel.
Much like with earlier generations, the _500 (9500, 9500T) is the i5 that Intel wants to sell in large quantities.
So if AMD decides to go for very aggressive pricing in this segment, Intel might let them have it.

That's just your bias, not matter-of-fact.

FYI, Ryzen 3000 is faster clock-for-clock, so a 4.1 GHz 3500 beats a 4.1 GHz 9400.

https://static.techspot.com/articles-info/1876/bench/CB20-multi.png

 

[HD64G]

Since the AM4 chipset will get even the Ryzen 4000 next year and will get the 16-core monster CPU in September, why anyone to buy any Intel CPU being close in both performance and price to a Ryzen 3000 one (apart from the top ones and those only for gaming at 240Hz and on +$1000 GPUs)? In VFM AMD wins easily both at APUs and CPUs on desktop especially since Ryzen launched. And now, servers, data centers and cloud is a Rome's field. Intel is clearly 2nd now in computing PCs anywhere (performace and efficiency wise) apart from mobile CPU market until Zen2 gets there too. Do your reality check people. Denial doesn't help to face reality eh?

 

[notb]

FYI, Ryzen 3000 is faster clock-for-clock, so a 4.1 GHz 3500 beats a 4.1 GHz 9400.

https://static.techspot.com/articles-info/1876/bench/CB20-multi.png

These clock-for-clock comparisons are pointless for real-life choices, but if it makes you happy...

1) All CPUs on this graph have HT/SMT, while the CPUs discussed here don't. This is important, because different software will show different utilization of SMT vs HT.
Check this: https://www.phoronix.com/scan.php?page=news_item&px=AMD-Ryzen-9-3900X-SMT-Perf
So if anything, I'd rather go for the single-thread Cinebench (which Ryzen also wins, so you're still happy)

2) Cinebench is a rendering benchmark, so is it a very good choice in case of 3500 vs 9400?
Will these CPUs be used for rendering? For any "workstation-type load" for that matter?
Maybe yes, but it's more likely that they'll be used for gaming. And in gaming, also clock-for-clock Intel still has the advantage.
Gaming comparisons are available in the same review:

4GHz CPU Battle: Ryzen 3900X vs. 3700X vs. Core i9-9900K

Expanding upon all the testing we performed in our day-one 3rd-gen Ryzen coverage, today we'll be running a clock-for-clock comparison benchmark. IPC can be a good indicator...

www.techspot.com

but you've decided to use Cinebench... and you call me biased...

3) Cinebench is known to give very high marks to Ryzen CPUs. I'm by no means suggesting that this engine is optimized for AMD or that AMD optimizes CPUs to perform well in a particular benchmark (this sounds unrealistic, right? => Volkswagen ;-) )
It's just a very lucky (for AMD) coincidence that the most popular benchmark works well on AMD's latest CPUs. Surely helps the sales.
But I'd not suggest buying a CPU based on that.

TechSpot also tested V-Ray, which is a lot more popular rendering engine. In this test 3700X beat 9900K by 4%. In Cinebench it's 13%.

What I should or should not do is immaterial and not the point. The point is about the article/comparison the OP linked to and it used the 9400, not the 9400F. If you feel Tom's Hardware should have used the F processors instead, your argument should be with them, not me.

You've followed their comparison and concluded that as a result you "don't have a problem with the Ryzen 5 3500's clocks". That's why I'm addressing you, not TPU.
So yes, you're comparing 3500 to 9400. And you're making the same mistake TH did.

As far as the other specs being "useless" well all I can say there is "yeah right".

When comparing any two similar, nearly identically matched products, it is about all of the minute specification details that differentiate (or compare similarly) the two. Not just one.

No, it's not. We should look at specs that are important for the target audience. Do you want to tell me that entry/mid-level desktop users care about cache size? Seriously?
At this point you should have mentioned that 3500 can be overclocked and it may even work with ECC RAM from time to time. ;-)

 

[tabascosauz]

These clock-for-clock comparisons are pointless for real-life choices, but if it makes you happy...

1) All CPUs on this graph have HT/SMT, while the CPUs discussed here don't. This is important, because different software will show different utilization of SMT vs HT.
Check this: https://www.phoronix.com/scan.php?page=news_item&px=AMD-Ryzen-9-3900X-SMT-Perf
So if anything, I'd rather go for the single-thread Cinebench (which Ryzen also wins, so you're still happy)

2) Cinebench is a rendering benchmark, so is it a very good choice in case of 3500 vs 9400?
Will these CPUs be used for rendering? For any "workstation-type load" for that matter?
Maybe yes, but it's more likely that they'll be used for gaming. And in gaming, also clock-for-clock Intel still has the advantage.
Gaming comparisons are available in the same review:

4GHz CPU Battle: Ryzen 3900X vs. 3700X vs. Core i9-9900K

Expanding upon all the testing we performed in our day-one 3rd-gen Ryzen coverage, today we'll be running a clock-for-clock comparison benchmark. IPC can be a good indicator...

www.techspot.com

but you've decided to use Cinebench... and you call me biased...

3) Cinebench is known to give very high marks to Ryzen CPUs. I'm by no means suggesting that this engine is optimized for AMD or that AMD optimizes CPUs to perform well in a particular benchmark (this sounds unrealistic, right? => Volkswagen ;-) )
It's just a very lucky (for AMD) coincidence that the most popular benchmark works well on AMD's latest CPUs. Surely helps the sales.
But I'd not suggest buying a CPU based on that.

TechSpot also tested V-Ray, which is a lot more popular rendering engine. In this test 3700X beat 9900K by 4%. In Cinebench it's 13%.

You've followed their comparison and concluded that as a result you "don't have a problem with the Ryzen 5 3500's clocks". That's why I'm addressing you, not TPU.
So yes, you're comparing 3500 to 9400. And you're making the same mistake TH did.

No, it's not. We should look at specs that are important for the target audience. Do you want to tell me that entry/mid-level desktop users care about cache size? Seriously?
At this point you should have mentioned that 3500 can be overclocked and it may even work with ECC RAM from time to time. ;-)

What are you on about? Clock for clock is pointless?

9400F is 4.1 boost and 2.9 sustained. 3500 should be 4.1 boost and 3.6 sustained. Ryzen has slightly higher IPC. Do I need to state the obvious here?

This is not a 3900X and 9900KF we're talking about.