From 83ea387ae1cac9e8563b3c941745f3c9d98d2f3b Mon Sep 17 00:00:00 2001 From: Tim Paine <3105306+timkpaine@users.noreply.github.com> Date: Tue, 22 Sep 2026 15:15:52 -0400 Subject: [PATCH] Rewrite the site copy in plain language The marketing copy read like it was generated: em dash runs, mirrored triads, negation stacks and sales lines. This states the same claims directly. No number, product name or performance claim changes; the three Playwright assertions that quoted the old wording are updated to the new. The Proof section still carries the temporal-workflow comparison that was withdrawn on 2026-08-03. That copy is untouched here and needs its own change once the replacement wording is decided. Signed-off-by: Tim Paine <3105306+timkpaine@users.noreply.github.com> --- MARKETING.md | 11 ++++----- src/lib/home/About.svelte | 43 ++++++++++++++++-------------------- src/lib/home/Contact.svelte | 9 ++++---- src/lib/home/Headline.svelte | 6 ++--- src/lib/home/Proof.svelte | 4 ++-- src/routes/+page.svelte | 2 +- tests/test.ts | 10 ++++----- 7 files changed, 37 insertions(+), 48 deletions(-) diff --git a/MARKETING.md b/MARKETING.md index 9dbcee4..3dc3dc5 100644 --- a/MARKETING.md +++ b/MARKETING.md @@ -11,13 +11,10 @@ maps supported Polars, dataframe, time-series, and domain operations into right-sized streaming hardware while unsupported work remains in familiar software. -Primary distinction: - -- not a fixed database appliance; -- not per-query synthesis in the runtime path; -- not a claim that every dataframe operation belongs on an FPGA; -- a reusable platform whose configurations improve around recurring workloads - and scale across device classes. +DAU is a reusable platform whose configurations improve around recurring +workloads and scale across device classes. It is not a fixed database +appliance, it does not synthesize hardware per query in the runtime path, and +it does not assume that every dataframe operation belongs on an FPGA. ## Search-intent clusters diff --git a/src/lib/home/About.svelte b/src/lib/home/About.svelte index 7ad8f2e..7be8652 100644 --- a/src/lib/home/About.svelte +++ b/src/lib/home/About.svelte @@ -1,17 +1,15 @@
-

- One platform, every scale -

+

The platform

Hardware shaped to your workload. Software that stays familiar.

- dau selectively accelerates the parts of Polars, dataframe, time-series, and analytical - workflows that benefit from streaming FPGA hardware. Reusable operations are composed into - right-sized configurations, while unsupported work remains in familiar software. The same - model scales from a personal accelerator to multi-card systems. + dau accelerates the parts of Polars, dataframe, time-series, and analytical workflows that + suit streaming FPGA hardware. Reusable operations compose into right-sized configurations; + unsupported work stays in software. The same model scales from a personal accelerator to + multi-card systems.

@@ -24,8 +22,7 @@

- Work stays expressed through the same software model as capacity, memory, and physical - placement change. + The software model stays the same as capacity, memory, and physical placement change.

@@ -136,8 +133,8 @@
- dpv1 is the first platform and where the approach was proven, dpv2 is the current platform - running today, and dpv3 is a planned design. + dpv1 was the first platform and where the approach was proven. dpv2 is the current + platform, running today. dpv3 is a planned design.
@@ -147,26 +144,24 @@

Personal computing

An accelerator beside you.

- An accessible accelerator for laptop and workstation development. Load configurations - matched to different analytical workflows without rebuilding applications around low-level - hardware APIs. + An accelerator for laptop and workstation development. Load a configuration that matches + the workflow you are running, without rebuilding the application around low-level hardware + APIs.

-

Portable · Flashable · Developer-friendly

+

Portable · Flashable

High-performance systems

-

- A fabric built around your data. -

+

Many cards, one workload.

Multiple accelerator cards partition data and workflows across independent local memory. The platform is designed to grow resident capacity and analytical throughput from a workstation to fabric-connected server arrays.

-

Parallel · Composable · Scalable

+

Parallel · Composable

@@ -201,11 +196,11 @@

The dau flywheel

- A configuration loop that improves around real work. + Configurations that improve as workloads are measured.

- Workload evidence guides what runs in hardware today and what configuration should be - built next. Every deployed design moves through the same verification path. + Measurements decide what runs in hardware today and what configuration gets built next. + Every deployed design goes through the same verification path.

@@ -245,8 +240,8 @@ diff --git a/src/lib/home/Contact.svelte b/src/lib/home/Contact.svelte index 2471a0e..12d6b97 100644 --- a/src/lib/home/Contact.svelte +++ b/src/lib/home/Contact.svelte @@ -12,15 +12,14 @@

- Build the next scale with us + Design partner program

Bring us the workload that should be faster.

- We are working with early users and design partners to shape dau hardware, software, and - deployment systems. Tell us what you compute, where it runs, and what performance would - change for you. + We are working with early users and design partners on dau hardware, software, and + deployment. Tell us what you compute, where it runs, and how fast it needs to be.

Become a design partner Follow the technology See the code on GitHub
diff --git a/src/lib/home/Headline.svelte b/src/lib/home/Headline.svelte index 1c7ce3e..b070e18 100644 --- a/src/lib/home/Headline.svelte +++ b/src/lib/home/Headline.svelte @@ -17,9 +17,9 @@ Data. Accelerated.

- dau maps high-value Polars, dataframe, time-series, and analytical operations into - reconfigurable FPGA dataflow while the rest stays in familiar software—from a personal - accelerator beside your laptop to data-center-scale systems. + dau maps selected Polars, dataframe, time-series, and analytical operations into + reconfigurable FPGA dataflow and leaves the rest in software. The same model runs on a + personal accelerator beside your laptop and on data-center-scale systems.

204K LUTs · 10 GB DDR3 · PCIe x8

- Two memory systems—2 GB onboard for bandwidth, 8 GB SODIMM for resident capacity—and eight - parallel lanes. Where the fused temporal workflow now beats the CPU baseline. + Two memory systems (2 GB onboard for bandwidth, 8 GB SODIMM for resident capacity) and + eight parallel lanes. Where the fused temporal workflow now beats the CPU baseline.

diff --git a/src/routes/+page.svelte b/src/routes/+page.svelte index 25070a1..b3031fd 100644 --- a/src/routes/+page.svelte +++ b/src/routes/+page.svelte @@ -24,7 +24,7 @@ diff --git a/tests/test.ts b/tests/test.ts index 668f093..855862d 100644 --- a/tests/test.ts +++ b/tests/test.ts @@ -6,21 +6,19 @@ test('home page presents both ends of the dau platform', async ({ page }) => { await expect(page.getByRole('heading', { level: 1, name: 'dau.' })).toBeVisible(); await expect(page.getByText('Data. Accelerated.')).toBeVisible(); await expect( - page.getByText('dau maps high-value Polars, dataframe, time-series', { exact: false }) + page.getByText('dau maps selected Polars, dataframe, time-series', { exact: false }) ).toBeVisible(); await expect(page.getByRole('heading', { name: 'An accelerator beside you.' })).toBeVisible(); - await expect( - page.getByRole('heading', { name: 'A fabric built around your data.' }) - ).toBeVisible(); + await expect(page.getByRole('heading', { name: 'Many cards, one workload.' })).toBeVisible(); await expect( page.getByRole('heading', { name: 'One platform, four deployment shapes.' }) ).toBeVisible(); await expect(page.getByText('A host-orchestrated dpv1 fleet')).toBeVisible(); await expect(page.getByText('Four dpv3 cards in a server')).toBeVisible(); await expect( - page.getByRole('heading', { name: 'A configuration loop that improves around real work.' }) + page.getByRole('heading', { name: 'Configurations that improve as workloads are measured.' }) ).toBeVisible(); - await expect(page.getByText('explicit fallback', { exact: false })).toBeVisible(); + await expect(page.getByText('The fallback is explicit', { exact: false })).toBeVisible(); await expect(page.getByText('dpv1 · Where it started')).toBeVisible(); await expect(page.getByText('dpv2 · Current platform')).toBeVisible(); await expect(page.getByText('dpv3 · Planned')).toBeVisible();