Intel Core i5-13400F

[Gica]

Overclocking

"Increasing BCLK beyond that limit isn't possible because Intel includes a BCLK counter in their CPU, which will shut off the CPU at 103 MHz and above."

I don't think it is possible with the 13 Series.

 

[Garrus]

The 13400F has too many cores for the 65W limit, it should have been slightly higher clocked and a 95W chip. It is a large $60-$80 price increase versus the old 12400F, and it doesn't have enough wattage to hit good clock speeds. Unlike the 12400F, I don't think the 13400F is a good buy. Get the 12600K or Ryzen 7600 (Just $20 CAD more), both are much better.

The price increase killed it.

 

[KrazyT]

this test is screwed, results for 7600 (X) are all wrong

Really ?
You should explain to us what's wrong ...

Nice little processor, efficient, still a bit pricey.
I was expecting better things in games compared to the 12400F, but slighty better in applications ...

For a gamer point of view, it's the kind of processor i would like to see more ...
You can save some $ and invest them in a better GC, a win-win scenario

There is something I still don't get ...
Why the results are often (slighty) worse when the power limit is removed ?
Shouldn't it be the opposite ?

 

[PenguinBelly]

There appears to be an error in the following chart. Or does 7600/X actually score that much lower than 7700/X?

 

[Garrus]

I'm not sure if it's actually that close. The 13400F comes with a stock cooler, whereas the 12600K does not. Getting a decent cooler for the 12600K will widen the price gap.

Arctic 17 CO for $15 is better than the stock cooler and more than enough.

There appears to be an error in the following chart. Or does 7600/X actually score that much lower than 7700/X?

wow

most be his default motherboard settings, my Gigabyte doesn't run the 7600X like that at all

C28 ram is $20 more and is very important for Ryzen also

I agree, use midboostit with Asus and $20 more ram and there is a massive gap between the 13400F and 7600 in gaming (>10%)

nice catch

 

[Sithaer]

4-5 years from now when it might actually matter I'll bet youll be able to find this or a i7 13700f for cheap.

Thats probably true but at that point I will just build yet another at the time current gen/more up to date budget-mid range system like I always do.
I hardly ever upgrade in the same socket, like the last time I did that was back in the AM2 days.

I had a AM4/B350 mobo for almost 4 years w/o ever upgrading my 1600X and then I sold off the mobo+cpu on the second hand market in early 2022. 'My mobo had no Bios support for Zen 3 at the time'

Also I would need a new mobo anyway for a i 7 or even a K i 5, my current crappy Prime is not suited for anything higher than a locked i 5. 'Hardware Unboxed tested this mobo and the VRM is pretty bad '

Second hand 12400F goes for about 140$ here so I guess even if I need an upgrade thats what I will do and then sell my 12100F.

 

[nomdeplume]

Huh, I expected this to crush the competition in efficiency when restricted to 65W and it doesn't. But I also expected it to lose badly when not restricted to 65W and that also didn't happen. Plusses and minuses ends up being pretty good.

But wow how much faster this is than my aging i5-8400, which is considerably slower than the 10400 bottoming out those performance charts.

It appears you confused what the OS handles with how the processor is designed to work. Key word being work. This was actually what impressed me the most out of the box with 8400 coming from 2400. Sure on paper it was close to being twice the processor. Just how advanced it was at avoiding getting bogged down, how eager to work, was surprising. Even against the last two gen of 4 core i5.

IF there was a $200 gpu well matched to 13400 I'd upgrade in a moment. Even a $200 mobo might be a stretch in that regard.


Thank you @W1zzard for this very informative review and @GerKNG for the equally appreciated real world input reinforcing what charts don't tell.

 

[W1zzard]

13500, 13600 and 13700 in these charts

Will buy them soon

"Increasing BCLK beyond that limit isn't possible because Intel includes a BCLK counter in their CPU, which will shut off the CPU at 103 MHz and above."

I don't think it is possible with the 13 Series.

Let me try

 

[Gica]

most be his default motherboard settings, my Gigabyte doesn't run the 7600X like that at all

C28 ram is $20 more and is very important for Ryzen also

I agree, use midboostit with Asus and $20 more ram and there is a massive gap between the 13400F and 7600 in gaming (>10%)

nice catch

With what video card? Don't tell me that a 7600X with steroids (more expensive motherboard and memory) will increase by ">10%" a 3070Ti or weaker.
The fine print in the reviews is that these budget processors are tested next to the king of video cards in gaming. 1500 hp in my Trabant!!!!

 

[W1zzard]

The 13400F has too many cores for the 65W limit, it should have been slightly higher clocked and a 95W chip

The "too many cores" statement isn't true, or we would see much better performance/power with the power limit removed. Higher clocks and thus higher voltage would have been good though, and with such an increase the more power that you mentioned would be useful, too.

There appears to be an error in the following chart. Or does 7600/X actually score that much lower than 7700/X?

Let me double check, this doesn't look right

edit: You are right, the 1T score should be closer to 1900. Will retest and see if anything else is affected, too.

"Increasing BCLK beyond that limit isn't possible because Intel includes a BCLK counter in their CPU, which will shut off the CPU at 103 MHz and above."

I don't think it is possible with the 13 Series.

https://imgur.com/vWXV5Vj

Same as before.. 102.9 works great, 103.0 doesnt POST

 

[Gica]

13500 + Z690 Gaming X with last F22 BIOS version. I also tried with F20, the first version with 13th support.
The motherboard lacks the BLCK adjustment menu in the BIOS.
XTU does not install, Easy Tune crashes. Something is not right. I just upgraded from W10 to W11 on this system with i5-10500, but the problem starts with the BIOS. With 10500 there were no problems to add that 2.8%.
I will contact them.

 

[Readlight]

What's the difference in UVD, quick syct video, pure video?

 

[JustBenching]

This i5 is a shame. Be it of the same arch as the higher-ranked 13th gen CPUs (y'know, it's de facto the locked out 12600KF at a lower frequency) and with no E-cores (it's utterly hard to find anyone who needs E-cores considering this budget area) but with higher clocks. Say, a classic 6/12 CPU with real Raptor Lake cores clocked at 4500 MHz in the all-core turbo mode, that would be an "I'm buying it immediately" gaming product.

And as of reality, it's a useless chunk of silicon and metal. Want more, get a 12600K and OC it. Want cheaper, get a 12600KF and don't OC it. Same level absurd as an i5-9400F and 11900 series (where are my cores #8 and #9!?).

You are so wrong about ecores in gaming you have no idea. I made lots of tests at cpu bound resolutions with ecores on and off, with ecores off game engines use HT and 1% lows take a dramatic hit. I have some videos posted on my channel showing exactly that. Ecores are really really nice to have

Huh, I expected this to crush the competition in efficiency when restricted to 65W and it doesn't. But I also expected it to lose badly when not restricted to 65W and that also didn't happen. Plusses and minuses ends up being pretty good.

But wow how much faster this is than my aging i5-8400, which is considerably slower than the 10400 bottoming out those performance charts.

It doesn't? It literally topped the mt efficiency chart, lol.

Intel cpus are the efficiency kings

 

[Wasteland]

https://ark.intel.com/content/www/u...600kf-processor-20m-cache-up-to-4-90-ghz.html
Up to DDR5 4800 MT/s
Up to DDR4 3200 MT/s

https://ark.intel.com/content/www/u...3400f-processor-20m-cache-up-to-4-60-ghz.html
Up to DDR5 4800 MT/s
Up to DDR4 3200 MT/s

Non-K Alder/Raptor Lake CPUs have a locked SA Voltage at ~0.9 Volts. So although it's true that the spec sheet shows the same memory support for e.g. the 12600k versus the 13400, in practice the 12600k has a higher tolerance for different memory configurations.

For example, the K variant will run DDR4-3600 in Gear 1 mode with no problem. Just flip the XMP toggle. The non-K variant will be unstable in the same configuration. Note too that this issue applies even on B-series motherboards; Intel actually muddied the waters here by marketing B-series boards as "memory overclockable" while largely glossing over the locked System Agent voltage on their non-K CPUs.

This memory thing isn't a big deal if you know about it, but it is worth mentioning.

 

[Ayhamb99]

So this is basically a watered-down 12600k with lower consumption, meh tbh, makes me feel better about my 12400f purchase in December, did not miss out on a lot in terms of gaming performance since i am using an RTX 2060. If i were to ever upgrade down the road a 13600k would suit me just fine but GPU first needs to be upgraded.


The 7600 and 7600X single threaded/gaming power consumption and efficiency though...... the hell happened there.

 

[Nostras]

I feel like pairing the 13400F with DDR5 is a bit of a mistake. No sane person should consider this setup.
Either you go cheap and do 13400F + DDR4 or the 5700X + DDR4 or you go bigger and do 13600KF + DDR5 or 7600X/7700X + DDR5.
Claiming that the 13400F beats Zen 3 is "technically" correct, but in practice this requires some very odd if not downright bad build choices.
Gamersnexus and HUB tested it as well, just like the 13500 and in general the conclusion I came to is simple:
- The 13400F is a pointless product as it typically gets outperformed by the 5700X, it only nudging a small victory in productivity
- The 13500 is an interesting workstation CPU on the cheap with a useful igpu (QS), but is bad product for gaming and a bit too close to the 13600K(F) in price
The 13400F and 13500 do have an advantage in platform upgradeability, but this becomes very questionable once you consider that you must buy a DDR4 set for this to make sense (which will kneecap the 13700 and 13900) and the fact that you must buy a quality motherboard that can deliver enough power without throttling (which conflicts with a "cheap" DDR4 board).

I do appreciate one review showing what the 13400F is "technically" capable of, but once you factor in platform costs as configured in this review it becomes kinda pointless.

Just my 2 cents.

The price of the 5700X is also lower than the 13400F which makes it even poorer value, but I can't fault the review for that. Pricing is local.

 

[Selaya]

but then you'll run into the 1:1 bottleneck/limit; 12th and 13th gen non-k have their SA voltages locked, which seriously limits their memory OC potential

at that point you're far better off buying a 5600 that can quite reliably run 3600, while the ADL parts are usually capped at 3200 (3600 memory should be as cheap as 3200 these days)
or, as i've suggested b4 - the 12600k - which as a k part has unlocked SA voltages, allowing you to run 3600 w/o a hitch.

this sku just makes little sense given the current pricing. now, if it were like $130 that'd be something else entirely but alas.

https://ark.intel.com/content/www/u...600kf-processor-20m-cache-up-to-4-90-ghz.html
Up to DDR5 4800 MT/s
Up to DDR4 3200 MT/s

https://ark.intel.com/content/www/u...3400f-processor-20m-cache-up-to-4-60-ghz.html
Up to DDR5 4800 MT/s
Up to DDR4 3200 MT/s

see the above explanation about locked SA voltages

 

[Macro Device]

with ecores off game engines use HT and 1% lows take a dramatic hit.

Did you test it on an i5-13400? Feels like you are counting wrong chickens considering i9-12900 test results are to apply to an i5.

My older RKL is really better in 1% lows whilst HT on but the CPU was damaged irreversibly rendering unable to be HT positive not going absurdly watt-hungry. In the 8c8t mode, 1% lows are considerably worse than they were in the 8c16t mode.

In other words, flick me the videos please.

 

[W1zzard]

The 7600 and 7600X single threaded/gaming power consumption and efficiency though...... the hell happened there.

Some error with Cinebench 1T results, I'm investigating right now

 

[JustBenching]

Did you test it on an i5-13400? Feels like you are counting wrong chickens considering i9-12900 test results are to apply to an i5.

My older RKL is really better in 1% lows whilst HT on but the CPU was damaged irreversibly rendering unable to be HT positive not going absurdly watt-hungry. In the 8c8t mode, 1% lows are considerably worse than they were in the 8c16t mode.

In other words, flick me the videos please.

Of course lows are better with HT on, but they are not better compared to actual physical cores like ecores

 

[Tahagomizer]

Good review. What grinds my gears in not quoting max all core turbo frequency anywhere. I fell into this trap with a 12400 I bought lately, it's sold as a 4.4GHz CPU but good luck seeing 4.4 GHz unless you start a single-threaded load and limit affinity to a single core. I've never seen it go beyond 4.2GHz for a split second and for any realistic load it's a 4GHz CPU. Even Intel seems to be ashamed of that, hiding actual frequencies from potential customers with the usual "up to" rubbish.
Otherwise a good CPU, just scammy marketing.

 

[Ayhamb99]

Some error with Cinebench 1T results, I'm investigating right now

Ah alright, As always great review and I appreciate all the work that you put into these reviews.

 

[W1zzard]

What grinds my gears in not quoting max all core turbo frequency anywhere

Intel Core i5-13400F Review - Force of Efficiency

The Intel Core i5-13400F is only $200, but comes with six Performance Cores and four E-Cores, bringing the total thread count to 16. Performance is good, as expected, but what's even more impressive is the energy efficiency as confirmed by the detailed testing in our review.

www.techpowerup.com

This is tested, not based on any claims or configs in the CPU

I think you're looking for a table that says "x cores = y MHz" ? That's some official spec and as you've noticed yourself mostly bs

 

[HD64G]

Some error with Cinebench 1T results, I'm investigating right now

Maybe comparing the 7600X review to the performance of it in this review of 13400F could help you point out what happened and make the correction easier.

 

[Imouto]

It doesn't? It literally topped the mt efficiency chart, lol.

Intel cpus are the efficiency kings

Because @W1zzard left the R9 7950X3D out. As per its own review:

See the i5 13400F included? And you accuse other people of fanboyism when it makes yourself blind.