AMD Ryzen 9 7945HX Beats Core i9-13950HX In Gaming Performance, Dragon Range Equipped Laptops Available Now

[T0@st]

AMD has announced the immediate availability of its Ryzen 7045 HX-series (Dragon Range) processors for high performance laptop devices. In a Youtube video released on March 10, AMD's Jason Banta has announced the availability of the world's most powerful mobile processor, the Ryzen 9 7945HX. He listed the OEM partners who have integrated the 7945HX into flagship level laptop models. He also declared that this range topping CPU is a competition beater. Gaming benchmark tests have demonstrated that the Ryzen 9 7945HX beats Intel's Raptor Lake Core i9-13950HX by an average margin of 10%.



The "Dragon Range" series is comprised of four processor models, the aforementioned Ryzen 9 7945HX is top dog, followed by the 12-core/24-thread Ryzen 9 7845HX, both featuring dual-CCD designs. These are trailed by the single-CCD models, the 8-core/16-thread Ryzen 7 7745HX, and the 6-core/12-thread Ryzen 5 7645HX. The Ryzen 9 7945HX is clocked at a 2.50 GHz base, with a maximum boost frequency of 5.40 GHz. The 7845HX has a 3.00 GHz base frequency, boosting up to 5.20 GHz. The 7745HX ticks at 3.60 GHz base, boosting up to 5.10 GHz. The 7645HX has the highest base frequency at 4.00 GHz, but only boots up to a maximum of 5.00 GHz.

Retail listings for laptops built using the Ryzen 9 7945HX have popped up on Newegg and Amazon. It seems that ASUS is the first company out of the gate to push product to market - with various configurations of its ROG range: Zephyrus Duo 16, Strix SCAR and Strix G17 models are available to purchase right now. At the current time, a basic G17 model will set you back $1799, and the top-of-the-line Zephyrus comes in at a wallet punishing $5145.



It will be interesting to see whether supply will meet demand, as AMD has been affected by supply chain and manufacturing issues in the recent past, especially during initial product launch periods.

View at TechPowerUp Main Site | Source

 

[Vader]

Nice to see laptops moving forward at such pace. Hopefully prices come down with time

 

[AnarchoPrimitiv]

Let's see if OEMs actually use these superior chips....I've got a feeling we don't see as many AMD laptops because AMD doesn't give enough bribes....I mean "development cost reimbursement"...to OEMs like Intel does. It's almost as if OEMs what make an AMD design unless they incur zero costs.

 

[Ivaroeines]

To me a TDP of 55W seem a bit to high to be in a laptop, of course they could undervolt and underclock the cpu/gpu, but i think AMD(possibly Intel too) need to do some serious work on power efficiency/low power, the gap ARM(and with them Apple) has on AMD in that respect will end all hopes AMD may have in the portable market segment.

 

[Wirko]

Meanwhile, the best integrated graphics processor in existence, the one in the 7940HS, remains unused because it's paired with a discrete GPU in all notebooks so far.

 

[neblogai]

These are fast CPUs, and efficient in heavy loads, but I do not like the fact that AMD is selling them before proper reviews. There are issues with the info provided thus far:
1) gaming performance, tested by Notebookcheck, was ~7% slower than 13950HX, not 10% faster, as claimed by AMD. I'm not sure what 13950HX laptop they picked as their competitor for their slides, but it had to be gimped one, in order to get these results that are most likely false.
2) every preview was boasting high power efficiency with load, but hiding battery life numbers at low load. Now the first review with that came out, and it is 3.5 hours of streaming video when set to be more power efficient (50% dimmed screen, iGPU use).

I'm an AMD fan, but I hate what their marketing people are doing recently, including this BS, and also claiming the 7800X3D is 20% faster than 13900K, as well as launching the 7900XT at $900 only to collect poor reviews (due to it being overpriced, when AMD knew well the performance and price of competitor 4070Ti already), and then have to drop the price to a more reasonable soon after.

 

[bonehead123]

Oh gee, moar miniscule incrementally-milked speed increases, with no other goal than to squeeze even moar $$ out of us for very little additional performance...

s.o.s.s... yada yada yada....

Can someone please wake me up when we can buy an 8GHZ, 35W, $600 lappy ?

TIA

 

[Leavenfish]

Oh gee, moar miniscule incrementally-milked speed increases, with no other goal than to squeeze even moar $$ out of us for very little additional performance...

s.o.s.s... yada yada yada....

Can someone please wake me up when we can buy an 8GHZ, 35W, $600 lappy ?

TIA

Enjoy your time in stasis and perhaps when you awake, Cancer will cured...war will be obsolete and we will have colonized Mars.

 

[Count von Schwalbe]

To me a TDP of 55W seem a bit to high to be in a laptop,

Choice, my friend. These are for performance segment laptops, ones designed to pack in the performance at the cost of everything else. Essentially a more portable desktop.

Phoenix is coming for the efficiency segment. The trouble with cutting the TDP of Dragon Range below 55W is that you begin to starve the cores of power, as the IF takes around ~20W. No doubt this could be reduced by cutting the IF speed, but RAM performance suffers.

Why are these here and Phoenix is not? These are simply the desktop processors on a different substrate. Phoenix is a completely different design, being fully monolithic.

No doubt Phoenix will also make it to the desktop, much later.

 

[PerfectWave]

"It will be interesting to see whether supply will meet demand, as AMD has been affected by supply chain and manufacturing issues in the recent past, especially during initial product launch periods." WHHAAAATTTTTTTT???????

 

[Garrus]

And the best part is the fastest CPU chip you can buy for a laptop is in a laptop that costs less than $2000, with an RTX 4060.

 

[mechtech]

Should have got a decent gaming laptop during the pandemic. Same ones now are 50% more, or out of stock.

 

[Denver]

These are fast CPUs, and efficient in heavy loads, but I do not like the fact that AMD is selling them before proper reviews. There are issues with the info provided thus far:
1) gaming performance, tested by Notebookcheck, was ~7% slower than 13950HX, not 10% faster, as claimed by AMD. I'm not sure what 13950HX laptop they picked as their competitor for their slides, but it had to be gimped one, in order to get these results that are most likely false.
2) every preview was boasting high power efficiency with load, but hiding battery life numbers at low load. Now the first review with that came out, and it is 3.5 hours of streaming video when set to be more power efficient (50% dimmed screen, iGPU use).

I'm an AMD fan, but I hate what their marketing people are doing recently, including this BS, and also claiming the 7800X3D is 20% faster than 13900K, as well as launching the 7900XT at $900 only to collect poor reviews (due to it being overpriced, when AMD knew well the performance and price of competitor 4070Ti already), and then have to drop the price to a more reasonable soon after.

There is a huge variation in performance between the same component on different laptops, CPU and GPU TDP configuration, cooling capacity, form factor etc...

The best comparison would be on the same laptop model, GPU, TDP and just changing the CPU.

 

[tabascosauz]

Choice, my friend. These are for performance segment laptops, ones designed to pack in the performance at the cost of everything else. Essentially a more portable desktop.

Phoenix is coming for the efficiency segment. The trouble with cutting the TDP of Dragon Range below 55W is that you begin to starve the cores of power, as the IF takes around ~20W. No doubt this could be reduced by cutting the IF speed, but RAM performance suffers.

Why are these here and Phoenix is not? These are simply the desktop processors on a different substrate. Phoenix is a completely different design, being fully monolithic.

No doubt Phoenix will also make it to the desktop, much later.

The new N6 IOD itself is not as bad as before, and laptops don't run nearly as high UCLK and Fabric so you can reduce power even further. The problem is not the IOD itself but the random 10W+ lost to the nether in all of AMD's chiplet designs, outside of core or SOC power. It was true for the CPUs and now is also true for Navi31, there's a lot of untouchable overhead and only AMD can do something about it.

No doubt AMD will bin the best dies for Dragon Range, but binning doesn't fix IFOP's problems. The 1CCD parts are getting better but most of it comes from lower IOD power. 2CCD is always fighting an uphill battle.

Obviously, if actually rendering or gaming then both Dragon Range and Raptor Lake laptops need to be plugged in, but if Dragon Range doesn't fundamentally do something different then it will also struggle with idle power on battery. It still is a laptop after all, gotta still be able to spend *some* time away from the plug......unless they ONLY want Clevo/Sager type design wins.

 

[Count von Schwalbe]

The problem is not the IOD itself but the random 10W+ lost to the nether in all of AMD's chiplet designs, outside of core or SOC power

This is the bit I was referring to.

It still is a laptop after all, gotta still be able to spend *some* time away from the plug......unless they ONLY want Clevo/Sager type design wins.

I have a laptop that approximates the type of thing this goes into. 100% designed to be used as a portable desktop, plug in and use. Battery is almost strictly to prevent having to shut down to move it.

I certainly wouldn't buy any laptop with this in it, if my goal was to be able to use it for emails in a coffee shop or airport. That is what the low TDP models are for - without dGPU.

 

[Minus Infinity]

Why does the 7645HX have the lowest boost clocks. Is that deliberate crippling to protect the "flagship" status of the 7945HX? Or is it just the best silicon goes from top to bottom?

Anyway these look great and Zen 4 has far better performance scaling at lower power than Raptor Lake. Keep to the 45-75W range and Zen 4 will kill RL.

Still I'm far more interested in Phoenix, 7740/7840H(S) with it's RDNA3 iGPU will do very nicely in a light and thin laptop.

 

[Dr. Dro]

Why does the 7645HX have the lowest boost clocks. Is that deliberate crippling to protect the "flagship" status of the 7945HX? Or is it just the best silicon goes from top to bottom?

Anyway these look great and Zen 4 has far better performance scaling at lower power than Raptor Lake. Keep to the 45-75W range and Zen 4 will kill RL.

Still I'm far more interested in Phoenix, 7740/7840H(S) with it's RDNA3 iGPU will do very nicely in a light and thin laptop.

It's a bit of both, but essentially, yes. It would not be a good outlook for AMD if the 7645HX-equipped laptops began putting frame rates that outperformed the 7945HX due to its ability to clock higher within the same power and thermal budget and most games matching perfectly to a 6-core processor. Even with the maximum boost clocks limited, the 1.5 GHz higher operating base might as well cause that to occur as it is.

Raptor Lake being 3 process nodes behind (Intel 7 is a rebrand of their 10ESF process) is really showing at both extremes, mobile (with Zen 4's impossibly good power efficiency squeezing Raptor Lake as a viable option away) and the ultra high performance desktop segment (13900KS x 7950X3D, with both tying up performance, each has its own bunch of wins but the Intel processor uses twice as much energy to do so).

 

[SL2]

These are fast CPUs, and efficient in heavy loads, but I do not like the fact that AMD is selling them before proper reviews. There are issues with the info provided thus far:
1) gaming performance, tested by Notebookcheck, was ~7% slower than 13950HX, not 10% faster, as claimed by AMD. I'm not sure what 13950HX laptop they picked as their competitor for their slides, but it had to be gimped one, in order to get these results that are most likely false.

You can't replace two laptops (the models AMD used) with two different ones (NBC) and expect the same outcome, that's pretty obvious. Also, the model is right there in the video.

AMD used a 17" Asus strix scar G17 that weights more than the dual display 16" Asus Zephyrus Duo 16 NBC used. It's pretty easy to guess which one's got the better cooling system, which will affect performance in the end.

The Zephyrus Duo 16 is probably a great machine overall, but nobody expects it to take the leads in any GPU benchmarks due to its smaller size and weight.

and also claiming the 7800X3D is 20% faster than 13900K,

This is all wrong. AMD said up to 20 % in games, which TPU's review already have confirmed. Yeah you have to guess which review I'm talking about.

Why does the 7645HX have the lowest boost clocks. Is that deliberate crippling to protect the "flagship" status of the 7945HX? Or is it just the best silicon goes from top to bottom?

That's pretty standard IMO.

 

[trsttte]

Oh my, 16 performance cores in a laptop, that'll surely be fun. Would love a workstation with one of those and without a dedicated gpu (i don't work with that), maybe dell or lenovo does one.

 

[watzupken]

This is the bit I was referring to.

I have a laptop that approximates the type of thing this goes into. 100% designed to be used as a portable desktop, plug in and use. Battery is almost strictly to prevent having to shut down to move it.

I certainly wouldn't buy any laptop with this in it, if my goal was to be able to use it for emails in a coffee shop or airport. That is what the low TDP models are for - without dGPU.

If you are intending to use it to check emails and surfing net, you are better off with a tablet with a keyboard cover. These sorts of processors are powerful, but battery life takes the back seat.

 

[R0H1T]

So AMD decided to bring a shotgun to a fist fight

 

[Xajel]

"It will be interesting to see whether supply will meet demand, as AMD has been affected by supply chain and manufacturing issues in the recent past, especially during initial product launch periods." WHHAAAATTTTTTTT???????

This is true actually,
In the past two generations, I've been waiting for an AMD laptop to be released, but it only went on sale in October/November, and reached my region 3-4 months later (1year after announcement), the actual announcement was in CES with expected release in March/April.
But the only AMD laptops that were actually launched in March/April were the gaming and low end laptops, I was waiting for a prosumer/business laptop mainly, to be more exact, either ThinkBook 16 Pro, or the ZenBook Pro 16/VivoBook Pro 16, but nothing.

The intel version of these laptops were announced in September and already available in October, why AMD laptops tends to get delayed by months or sometimes a year in some regions !!

Sadly, this has been the case for a few generations now, not just 1-2. AMD seemed to be ordering low quantities to not over supply, but kept undersupplying the market each year and they just can't help to see what is the issue is. This not to mention they keep releasing old architectures with new branding, this already consume a lot of production/packaging time better tailored for more recent architectures. I know this is what some OEMs want (low prices for last gen.) but can't they just release lower tiers of the new gen in the same time, or at least soon enough?

 

[medi01]

Only if they'd pair it with something OTHER than overpriced green GPUs...

 

[A&P211]

Oh my, 16 performance cores in a laptop, that'll surely be fun. Would love a workstation with one of those and without a dedicated gpu (i don't work with that), maybe dell or lenovo does one.

AMD doesn't have performance cores, you are thinking of intel. Amd has normal cores.

 

[trsttte]

AMD doesn't have performance cores, you are thinking of intel. Amd has normal cores.

AMD only has performance cores - that's what's normal, we already had 16 cores on a laptop with intel shenanigans of 8p + 8E, but now we get a real 16 core solution - maybe, amd laptops are less common in general let alone workstation ones.