I have NON-NDA access to Ryzen with Vega..

Doesn't the Mobile R7 beat an MX150?


Not the currently-shipping 2500U mentioned on the previous page, that's actually priced competitively. The cheapest fully-unlocked 2700U laptops start at $950,and that's GTX 1050 territory. When it ships, anyway.

https://www.amazon.com/Acer-Swift-Ryzen-Windows-SF315-41-R6J9/dp/B078B5WZZ5

My review data for the more reasonable-priced 2500U is sourced from here:

https://www.notebookcheck.net/HP-En...00U-Radeon-Vega-8-Laptop-Review.266614.0.html

https://www.notebookcheck.net/HP-Sp...-8550U-MX150-Convertible-Review.262442.0.html

Just comparing like GPU to like GPU, there are three comparable benchmarks:
These include low and high, to show you're not CPU-limited.

Bioshock Infinite:
2500U, Low = 114, Ultra = 20
MX150, Low = 201, Ultra = 34.8

The Witcher 3:
2500U, Low = 42.3, High = 14
MX150, Low = 68.3, High = 22.8

Rise of the Tomb Raider
2500U, Low = 49.3, High = 15.4
MX150, Low = 72.3, High = 23.1

With the more modern titles, that's a 50-60% performance difference when you stress the GPU at highest playable settings, where the CPU plays no part. The 2700U will have a tough time bridging that massive gap, with only 25% more compute units unlocked.

The desktop 2400G parts have an easier time exceeding the MX150 because they have twice the power budget to work with. But if the notebooks used that, the battery would run flat in half hour.

An example of a GTX 1050 thin laptop that competes/exceeds that Acer Swift 3:

https://www.newegg.com/Product/Product.aspx?Item=N82E16834234645
 
Last edited:
As an Amazon Associate, HardForum may earn from qualifying purchases.
The only problem is,AMD doesn't get any better efficiency in this combined package. That HP notebook with Vega APU has lower performance than the MX150, which is available in similar-specced and similar -priced systems.

https://www.newegg.com/Product/Prod...154&cm_re=mx150_laptop-_-34-316-154-_-Product


The AMD GPU uses the same amount of power as the MX150 notebooks as well. The desktop version offers much faster performance, but at a doubling of power consumption.

AMD has a much better chance of making inroads with the discrete Intel Kaby G parts on notebooks. Every high-performance Raven Ridge notebook with dual-channel DDR4 will have the same large PCB as one containing single channel DDR4 and 64-bit GDDR5. And there's no price, performance or power advantage.

you might see these in non ultrabooks just like intel's HQ processors. they would probably be lower gpu clocked versions of the 2400g.
 
Last edited:
Not the currently-shipping 2500U mentioned on the previous page, that's actually priced competitively. The cheapest fully-unlocked 2700U laptops start at $950,and that's GTX 1050 territory. When it ships, anyway.

https://www.amazon.com/Acer-Swift-Ryzen-Windows-SF315-41-R6J9/dp/B078B5WZZ5

My review data for the more reasonable-priced 2500U is sourced from here:

https://www.notebookcheck.net/HP-En...00U-Radeon-Vega-8-Laptop-Review.266614.0.html

https://www.notebookcheck.net/HP-Sp...-8550U-MX150-Convertible-Review.262442.0.html

Just comparing like GPU to like GPU, there are three comparable benchmarks:
These include low and high, to show you're not CPU-limited.

Bioshock Infinite:
2500U, Low = 114, Ultra = 20
MX150, Low = 201, Ultra = 34.8

The Witcher 3:
2500U, Low = 42.3, High = 14
MX150, Low = 68.3, High = 22.8

Rise of the Tomb Raider
2500U, Low = 49.3, High = 15.4
MX150, Low = 72.3, High = 23.1

With the more modern titles, that's a 50-60% performance difference when you stress the GPU at highest playable settings, where the CPU plays no part. The 2700U will have a tough time bridging that massive gap, with only 25% more compute units unlocked.

The desktop 2400G parts have an easier time exceeding the MX150 because they have twice the power budget to work with. But if the notebooks used that, the battery would run flat in half hour.

An example of a GTX 1050 thin laptop that competes/exceeds that Acer Swift 3:

https://www.newegg.com/Product/Product.aspx?Item=N82E16834234645

There are huge price differences between the two AMD driven units and the Intel/Nvidia one, less than half the price. The MX150 is also a much bigger GPU with its own power draw so the power reflected is AMD's total package power, but the MX is on its own. If battery life is the way of the game the 87XX and MX150 will probably kill a battery fast, provided the 87XX hits its boost speeds without serious throttling.
 
As an Amazon Associate, HardForum may earn from qualifying purchases.
Yes, a $20 markup over MSRP. It's available at MSRP on amazon.

I looked about an hour ago and didn't see it. Thanks for the heads up.

Edit: What is it under? I can't find it searching for Ryzen 5 or 2400g
 
As mentioned, Newegg is showing stock of these APUs but DAMN they are way jacked up, like +$30 on the Ryzen 3 and +$20 (IIRC) on the Ryzen 5:

AMD RYZEN 3 2200G Quad-Core 3.5 GHz (3.7 GHz Turbo) Socket AM4 65W YD2200C5FBBOX Desktop Processor
$129.99 (supposed AMD SRP "is"/"was" $99...
https://www.newegg.com/Product/Product.aspx?Item=N82E16819113481&ignorebbr=1

AMD RYZEN 5 2400G Quad-Core 3.6 GHz (3.9 GHz Turbo) Socket AM4 65W YD2400C5FBBOX Desktop Processor
$189.99 (supposed AMD SRP "is"/"was" $169...
https://www.newegg.com/Product/Product.aspx?Item=N82E16819113480&ignorebbr=1

So pretty sure Newegg is pulling the typical bend-over-the-first-buyers tactics vendors like that tend to do. Or AMD changed their SRP on these. And yes I know the S means "suggested."
 
As mentioned, Newegg is showing stock of these APUs but DAMN they are way jacked up, like +$30 on the Ryzen 3 and +$20 (IIRC) on the Ryzen 5:

AMD RYZEN 3 2200G Quad-Core 3.5 GHz (3.7 GHz Turbo) Socket AM4 65W YD2200C5FBBOX Desktop Processor
$129.99 (supposed AMD SRP "is"/"was" $99...
https://www.newegg.com/Product/Product.aspx?Item=N82E16819113481&ignorebbr=1

AMD RYZEN 5 2400G Quad-Core 3.6 GHz (3.9 GHz Turbo) Socket AM4 65W YD2400C5FBBOX Desktop Processor
$189.99 (supposed AMD SRP "is"/"was" $169...
https://www.newegg.com/Product/Product.aspx?Item=N82E16819113480&ignorebbr=1

So pretty sure Newegg is pulling the typical bend-over-the-first-buyers tactics vendors like that tend to do. Or AMD changed their SRP on these. And yes I know the S means "suggested."

typical newegg.. amazon has it for the $170 MSRP.
 
typical newegg.. amazon has it for the $170 MSRP.
Yup, just checked and Amazon indeed has them both @ SEP/SRP/whateverz price. Shame on Newegg. Do they PM Amazon? Someone needs to finger waggle the folks at Newegg. LOL
 
numbers don't look too bad.. for comparison sakes i score 101.69 with my 1050ti(system in sig) so 70 i'd guess the vega 11 sits right around the RX 550/560.


you might see these in non ultrabooks just like intel's HQ processors. they would probably be lower gpu clocked versions of the 2400g.


They already compare with the 15w Intel CPUs with MX150 in terms of power consumption. Adding just those ex

There is no more room to grow unless you put it in a desktop.
There are huge price differences between the two AMD driven units and the Intel/Nvidia one, less than half the price. The MX150 is also a much bigger GPU with its own power draw so the power reflected is AMD's total package power, but the MX is on its own. If battery life is the way of the game the 87XX and MX150 will probably kill a battery fast, provided the 87XX hits its boost speeds without serious throttling.


From the Notebook Check review:

The Ryzen-based HP is roughly comparable to notebooks powered by ULV Intel CPUs and entry-level Nvidia GPUs. The Xiaomi Mi Pro is perhaps the best example of this as it carries the i5-8350U CPU, MX150 GPU, and the same size 1080p IPS display as our Ryzen HP. When subjected to average (3DMark06) and Witcher 3 loads, both systems are nearly identical at 46 W to 51 W each.

Ryzen the processor is power-efficient, and performs well against Intel. But APU Vega is just as inefficient as normal Vega.

So you still pay the same power price as a SEPARATE APU in the MX150 for 50% less performance. And the price premium you pay for that Raven Ridge puts it in exactly the same price range as this puppy with eighth-gen i5 and MX150:

https://www.amazon.com/Acer-Aspire-i5-8250U-GeForce-E5-576G-5762/dp/B075FLBJV7

Tell me again how Raven Ridge is good value? OR more efficient? OR better-performing? I just shot down all three in just one post.

You can't make the installation space smaller (requires 128-bit DDR4 to power it), so the notebook PCB size will be the same as a 64-bit CPU plus 64-bit GDDR5 GPU. So it doesn't have the space advantage of Kaby G.
 
Last edited:
As an Amazon Associate, HardForum may earn from qualifying purchases.
I feel like this is finally capturing what AMD was planning on Llano to accomplish so long ago :p

I agree. Its just i sometimes see people pushing these builds were you
http://www.legitreviews.com/amd-ryzen-5-2400g-mining-performance-nicehash-xmr-stak_202662

Only ~260H/s for both the GPU/CPU, so these should be readily available as they aren't all that valuable to miners.

My FX-6300 does 270h/s. At this point mining profitability is already in the toilet. even my ryzen 7 1700 is doing 550h/sec and gets 75 cents a day or less.
 
They already compare with the 15w Intel CPUs with MX150 in terms of power consumption. Adding just those ex

There is no more room to grow unless you put it in a desktop.



From the Notebook Check review:



Ryzen the processor is power-efficient, and performs well against Intel. But APU Vega is just as inefficient as normal Vega.

So you still pay the same power price as a SEPARATE APU in the MX150 for 50% less performance. And the price premium you pay for that Raven Ridge puts it in exactly the same price range as this puppy with eighth-gen i5 and MX150:

https://www.amazon.com/Acer-Aspire-i5-8250U-GeForce-E5-576G-5762/dp/B075FLBJV7

Tell me again how Raven Ridge is good value? OR more efficient? OR better-performing? I just shot down all three in just one post.

You can't make the installation space smaller (requires 128-bit DDR4 to power it), so the notebook PCB size will be the same as a 64-bit CPU plus 64-bit GDDR5 GPU. So it doesn't have the space advantage of Kaby G.

Based on other site's testing of that HP, the screen appears to draw a lot of power compared to other similar units. The Intel version (envy x360 m6) has similar pedestrian battery life, so I think your comparison (or rather Notebook Check's) is faulty. I'd wait for the Acer Swift 3 reviews.
 
As an Amazon Associate, HardForum may earn from qualifying purchases.
Based on other site's testing of that HP, the screen appears to draw a lot of power compared to other similar units. The Intel version (envy x360 m6) has similar pedestrian battery life, so I think your comparison (or rather Notebook Check's) is faulty. I'd wait for the Acer Swift 3 reviews.

Fair. But no 15" notebook screen in this life burns more than 10-12w. A difference of 5w between two screens makes all the difference in the world on WIFI Browsing BATTERY TESTS, but is lost in the noise when the rest of the laptop is fully-stressed.

Are you going to proclaim that 5w difference between two 60w gaming laptops is really going to put it in another class? That's the weakest claim I've read so far in this thread.

Screen power differences alone are not going to change the load power significantly. It's 10% (or less) of the load power at that point.

There are many differences between each laptop out there, even if they have the same specs. BUT OVERALL, the peak performance and peak power consumption very rarely varies by more than 10%.
 
Last edited:
Newegg dropped their inflated prices down to Amazon prices, aka where they should have been prices.
 
Fair. But no 15" notebook screen in this life burns more than 10-12w. A difference of 5w between two screens makes all the difference in the world on WIFI Browsing BATTERY TESTS, but is lost in the noise when the rest of the laptop is fully-stressed.

Are you going to proclaim that 5w difference between two 60w gaming laptops is really going to put it in another class? That's the weakest claim I've read so far in this thread.

Screen power differences alone are not going to change the load power significantly. It's 10% of the load power at that point.

There are many differences between each laptop out there, even if they have the same specs. BUT OVERALL, the peak performance and peak power consumption very rarely varies by more than 10%.

No, but it's 5w less than the intel/mx150 combo already, and another 5w would make it ~20% less, which is significant in a laptop. I don't know that's the case, but neither do you as it's a faulty comparison. In at least one review that they matched screen brightness, they had to pump the brightness of the HP screen all the way up to get it to 100cd/mm2 where the other laptops were much less. My point was that a better comparison will be the Acer Swift 3 sf15t-51 vs sf15t-41, which is AMD vs Intel straight up with the rest of the systems the same. I haven't seen it yet, but I'm sure it will be coming.

I also disagree that a 2nd ram channel is the same packaging and engineering effort as a dgpu and most of the intels are dual channel anyway, including the XIaomi you linked, so it's definitely a larger engineering and packaging effort to pack a dgpu in given that fact. The amd apu should be a cheaper overall solution than dual channel intel with dgpu by a decent amount.
 
Fine, you can wait for the $950 Acer 3. Meanwhile, the rest of the world will buy that $600 Acer I just linked, and enjoy higher performance and lower price.

Even if what you're imagining is true (10w difference because HP is stupid), 50w versus 60w is still high enough power consumption that you can't even get two hours out of the battery when gaming. So it's a plugged-in-or-bust gaming machine either way.

Your 10w imagined power consumption difference doesn't put the AMD chip in another class of notebook (you'd have to cut power by half to do that), and it's vastly slower and more expensive than it's peers.

A 30w entry-level gaming notebook with MX150 performance would be something. But 50-60w is just treading water with Nvidia. Lower performance sinks that ship.
 
Last edited:
Fine, you can wait for the Acer 3. Meanwhile, the rest of the world will buy that $600 Acer I just linked, and enjoy higher performance and lower price.

Even if what you're imagining is true (10w difference because HP is stupid), 50w versus 60w is still high enough power consumption that you can't even get two hours out of the battery when gaming. So it's a plugged-in-or-bust gaming machine either way.

Your 10w imagined power consumption difference doesn't put the AMD chip in another class of notebook, and it's vastly slower than it's peers.

That doesn't play in the same arena, it's a pound and a half more than the Swift 3. Different markets, here's the comparable Intel Acer https://www.amazon.com/dp/B0746P25QX/ref=psdc_13896615011_t4_B075FLBJV7. But yeah, at $150 less, I'd be inclined to go Intel/mx150 without a direct comparison.

As to performance, the 2700u is around the same as the mx150 intel combo, so if it uses 20% less power it's a winner at the same price. At $150 more, it's not, I agree. The 2500u Swift 3 is $50 less, so that may be a tougher decision. Less performance but less cost and power consumption (possibly) in thin and light. Probably go for the mx150, though.
 
As an Amazon Associate, HardForum may earn from qualifying purchases.
The 2700U will NOT match the performance of the MX150. The clocks of the MX150 are sightly higher than the GT 1030.

It's not fully-unlocked like the 2400G, and there's 10-20% lower clocks as well. I'd expect it to be 20% slower than the 2400G, which puts it MAYBE matching the MX150 in optimized games, and being 20% behind in everything else.
 
It also depends on whether it's matched to the 8250u (like the Acer I linked) or the 8550u, which raises the price another $150-$175. So yeah, for outright performance and price/performance, I think the 8550u-mx150 combo will be best, but I also think that combo will use significantly more power than a 15w 2700u implementation (which is what the Swift 3 is, HP is a 25w implementation), as in double power consumption for around 25% performance gain. Horses for courses I guess.
 
Back
Top